Southern Cross University

ePublications@SCU
School of Environment, Science and Engineering Papers 1995 School of Environment, Science and Engineering

The mangrove vegetation of the Atlantic Coast of Africa: a review

Peter Saenger
Southern Cross University

M F. Bellan

Publication details
Saenger, P & Bellan, MF 1995, The mangrove vegetation of the Atlantic Coast of Africa: a review, Universit de Toulouse, Toulouse, France. Copyright P Saenger 1995

ePublications@SCU is an electronic repository administered by Southern Cross University Library. Its goal is to capture and preserve the intellectual output of Southern Cross University authors and researchers, and to increase visibility and impact through open access to researchers around the world. For further information please contact epubs@scu.edu.au.

The Mangrove Vegetation of the Atlantic Coast of Africa - A Review P. Saenger1 and M.F. Bellan2 for Coastal Management Southern Cross University Lismore, NSW 2480, Australia 2Laboratoire dEcologie Terrestre CNRS/UPS BP 4403 314 Toulouse Cedex, France.
"In every ... direction you will see the apparently endless walls of mangrove, unvarying in colour, unvarying in form, unvarying in height, save from perspective. Beneath and between you and them lie the rotting mud waters of Bonny River, and away up and down river, miles of rotting mud waters fringed with walls of rotting mud mangrove-swamp." Mary Kingsley (1897)
1Centre

Introduction The mangroves of West Africa are reasonably well-described floristically and ecologically, albeit in a widely scattered literature (Jackson, 1985), and they have recently been reviewed from a functional perspective (John and Lawson, 1990). However, a wider review of the mangrove vegetation along the entire Atlantic coast of Africa is justified to put these systems into a continental perspective and, thus, provide a basis for a regional comparison with the mangroves of the Atlantic coast of the Americas and elsewhere. In terms of what are generally regarded as mangroves (Barth, 1982; Saenger et al., 1983; Tomlinson, 1986), there are only six indigenous and one introduced woody species on the Atlantic coast of Africa (Wilcox, 1985). Clearly, this is a species-poor tropical plant assemblage, even if the associated species are considered. Apart from the intrinsic interest in such an ecosystem, questions as to the underlying causes of this impoverishment should provide some insights into the general understanding of mangrove systems. Present Mangrove Setting The Atlantic coast of Africa (from Gibraltar to the Cape of Good Hope) is approximately 14,040km long and it ranges from 36N to a latitude of 35S (fig. 1). Not surprisingly, it provides a large variety of coastal types in a range of climatic settings. Ocean Currents & Tidal Characteristics Several major ocean currents influence the western coast of Africa (Longhurst, 1962). Two cold currents, the Canary Current in the northern

hemisphere and the Benguela Current in the southern hemisphere, move towards the equator squeezing, as it were, the region of warm water on the coast into a narrower belt than would be expected from a consideration of latitude alone (Lawson, 1966). As they approach the equator, these two currents turn westward and are then known as the North and South Equatorial currents respectively. Between them, the warmer Counter Equatorial current flows in the opposite direction towards the coast, moving northwards during the northern winter. Like the Counter Equatorial current, the Guinea current flows in an easterly direction travelling along the length of the Guinea coast and turning south in the Bay of Bonny, joining the westward flowing South Equatorial current (fig. 2). Another feature of importance is the phenomenon of 'upwelling' of cold water which takes place along the coastline of the Gulf of Guinea. This causes short-term temperature drops and brings increased nutrients to the surface in June-August. Its influence is pronounced from western Nigeria to Cte d'Ivoire but it is much reduced west of there. Tides on the Atlantic coast of Africa are of the semi-diurnal type with ranges that are locally variable but generally small (Table 1). Atlantic swells are ubiquitous along most of the unprotected western coast of Africa and, consequently, it is only in semi-enclosed bays and in lagoons and estuaries, that conditions can be described as sheltered. As the amplitude of the waves often exceeds the range of the tides, wave action is relatively more important in determining the presence and vertical extent of mangrove vegetation on the open shoreline. In estuaries and lagoons, highly variable hydrodynamic conditions prevail and it is difficult to generalize. Rainfall & Temperature Ombrothermic diagrams, after Walter and Lieth (1960), are shown (fig. 3) for sixteen climatic stations from western Africa. These diagrams clearly indicate that much of the western coast of Africa is uniformly mild to warm throughout the year but that there are large differences in total annual rainfall. In the north, coastal Mauritania and northern Senegal are arid (911 dry months) while similar conditions prevail in the south, largely as a result of the cold offshore watermasses. In between these arid zones is situated the humid tropical zone with very high rainfalls in Guinea, Guinea-Bissau and Liberia as well as the head of the Gulf of Guinea (Nigeria, Cameroon, Equatorial Guinea and Gabon). The areas between the humid zones have an annual rainfall between 10002000mm punctuated by two short dry periods (Cte d'Ivoire, Ghana, Togo and Benin). It should be noted, however, that along the entire coastline, maximal rainfall generally coincides with the warmer months. Unlike many mangrove coastlines elsewhere, cyclone activity is absent on the Atlantic coast of Africa (Anthony, 1989).

Geomorphology, Physiography and Hydrology According to Lang and Paradis (1977; 1984), the last glaciation around 18,000 y. BP resulted in a lowering of sealevels of about 100-150m along the entire Gulf of Guinea coastline. Beginning around 10,000 y. BP, sea level rose again due to the melting of the glaciers, and reached a new maximum between 6,500 and 4,500 y. BP during the Nouakchottian transgression (Anthony, 1989). All valleys and river mouths were flooded in Senegal, Gambia, Guinea, Cte d'Ivoire and Gabon, forming rias still common today. Many interdunal areas in Mauritania and elsewhere were also flooded (up to 250km inland) and occupied by dense vegetation as a result of the humid conditions which accompanied this transgression (Faure and Hebrard, 1977; Elouard, 1968; Gowthorpe and Lamarche, 1993). The Nouakchottian transgression also left sand deposits considerable distances inland, many of which still exist today and comprise the landward margins of the coastal plains as, for example, in Cte d'Ivoire, Ghana, Benin, Sao Tome and Angola. Towards the end of the Nouakchottian transgression, between 4,800-4,200 y. BP, a north-south (along the Senegal to Guinea coast) and an east-west (along the Gulf of Guinea) littoral drift became established which gradually caused littoral sand-barriers to close the gulfs and rias that had been formed. This phenomenon was progressive, starting in Mauritania (4,000 y. BP), then Senegal (3,900-3,200 y. BP) and subsequently in Guinea, Cte d'Ivoire and Benin. By 3000 y. BP, for example, the mouth of the Casamance in Senegal was transformed into a large lagoon where mangroves had become established. This sand-barrier formation was further consolidated by the small Taffolian regression of about 1m at around 3,000 y. BP. By about 2,000-1,500 y. BP the Atlantic coastline of Africa had largely developed its present configuration and the climate had reverted abruptly from a humid type to the modern semi-arid climate without any transitional stages (Delibrias et al., 1976; Kalck, 1978; Barusseau et al., 1986; Lezines, 1988). Over the last 1,500 years, sealevel fluctuations have been small and difficult to establish with certainty, although sealevel rises of between 1-3.4mm/y have been reported from this region (Verstraete, 1989; Emery and Aubrey, 1991). As a consequence of these sealevel and climate changes, most of the mangrove vegetation has contracted significantly along the west coast of Africa, with the remnants along parts of the coastline (e.g. Senegal, Cte d'Ivoire, Ghana, Togo, Benin and western Nigeria) characteristically in 'physiographic refugias' i.e. semi-enclosed coastal lagoons or embayments, generally with constrained tidal exchange and with limited (and markedly seasonal) freshwater input. Additional 'physiographic refugias' can be recognized where offshore sandbanks (e.g. Guinea), coral communities (e.g. Limb, Cameroon) or coral rubble banks (e.g. Liberia, Nigeria) may protect

the shoreline from wave action which, in turn, allows frontal mangroves to develop. However, there are also some large deltas on this coastline which constitute 'salinity refugias'. Mangroves are associated with these river deltas particularly the larger ones such as the Gambia, Volta, Niger, Rio Muni, Ogoou, and Zaire; these settings are characterized by more or less continuous freshwater input, free tidal exchange as well as the abundant deposition of alluvial sediments. These areas can be further subdivided into estuarine (polyhaline) or fluvial (mesohaline) areas depending on the prevailing salinity regime. Mangrove Types As already noted by Grewe in 1941, open shoreline (frontal) mangroves are rare along the Atlantic coast because of the absence of fringing reefs - and thus the high energy exposure at most locations. Given the coastal settings described above, it appears that four mangrove types can be recognized on this coastline (Table 2). From Cape Verde to Liberia the coast is notched by deep-branching estuaries with low-silted sandbanks. The coastline east of Liberia to the Nigerian-Cameroon border is low and lined with barrier islands which enclose lagoons stretching parallel to the beach. Behind these lagoons a number of rivers empty. Thus, from Mauritania to Liberia, mangroves are predominantly deltaic (estuarine and fluvial) or frontal (Anthony, 1989) while all mangroves in Benin are lagoonal as there are no deltas and the open shoreline is sandy and of relatively high energy (Gaillard et al., 1982; Baglo, 1989). Along the Gulf of Guinea coast, mangroves are largely lagoonal although a few deltas (e.g. Mono and Volta Rivers) occur. In Nigeria, where there are 36 estuaries which may be bar-built (e.g. Lagos Lagoon), drowned river valleys (e.g. Kwa Ibo) and river deltas (e.g. Niger delta), mangroves are essentially estuarine (Amadi, 1991; Isebor, 1993). These differing settings are also reflected in the breadth of mangrove vegetation along the coastlines as discussed under the 'Extent of Mangrove Vegetation' below. Floristics The early literature concerning the mangrove vegetation of the Atlantic coast of Africa failed to distinguish the three species of Rhizophora (e.g. Pellegrin, 1952) and it was not until Keay (1953) clarified this situation, that reliable floristic data became available for analysis. As Breteler (1969) had suggested for the Atlantic Rhizophoras, Wilcox (1985) has indicated that R. harrisonii might be a hybrid of R. racemosa and R. mangle. This is supported by intermediate morphological features (such as lengths of peduncle and pedicels and the number of flowers in the inflorescence), the low viability of R. harrisonii pollen (Wilcox, 1985; Breteler,

1969) and the poor fruit set of this species (Keay, 1953; Savory, 1953; Breteler, 1969). Avicennia germinans has a similarly long synonymy (including a period when it was considered as an endemic species - A. africanum) before it became generally accepted that there were no significant morphological differences between the American and African populations (Compre, 1963). Given the current taxonomic status, there are six indigenous and one (Nypa fruticans) introduced species of mangroves on the African coast. The records of Osbornia octodonta for western Africa by Chapman (1977) and Barth (1982) are undoubtedly based on locality confusion. Even when associated species are added to this list (Table 3), only 12 families constitute this widespread vegetation type. All of these families have counterpart species or genera on the American Atlantic coasts and the majority have counterparts along the eastern coast of Africa. Vegetational History Based on pollen analysis of sediments from the Niger delta (Sowunmi, 1981; 1986), offshore cores (Rossignol-Strick and Duzer, 1979) and peat deposits (Michel and Assemien, 1970; Assemien, 1971), the extent and floristic composition of the mangrove vegetation of western Africa has altered dramatically in conjunction with changing climatic conditions and palaeosealevels. During the Palaeocene (63-55 m.y. BP) when the proto-Atlantic was a narrow waterway and western Africa was experiencing a 'wet' phase, there was no mangrove vegetation in the Niger delta but extensive estuarine swamp communities dominated by Nypa. The earliest Nypa pollen are recorded from Brazil (Hutchings and Saenger, 1987) and this genus apparently evolved in, and spread throughout, the proto-Atlantic at this time. This Nypa dominated swamp community remained during the Eocene when sealevels fell and drier conditions prevailed. Two additional genera of palms, Mauritia now confined to the Americas and Kentia, now only occurring in Australia and New Zealand, were also common members of these estuarine swamps of the Niger delta during this period. Towards the end of the Eocene when seasonally dry conditions were widespread in the region, Nypa, Mauritia and Kentia declined in abundance and finally disappeared altogether in the early Miocene (24 m.y. BP) incidentally at the same time Nypa disappeared from the fossil record of Venezuela. These disappearances coincided with the sudden and predominating appearance of Rhizophora throughout the region - a situation which, with minor fluctuations, remains to the present. Thus, during the Inchirian transgression (c. 35,000 y. BP) there was a peak in Rhizophora pollen, as again during the Nouakchottian transgression around 5,500 y. BP (Sowunmi, 1986).

Mangroves were widespread during the Nouakchottian transgression based on pollen (Assemien, 1969) and other fossils dated around 5,700 to 5,300 y. BP (Tastet, 1981; Lezines, 1988; Gowthorpe and Lamarche, 1993). The inland occurrence of Avicennia germinans (accompanied by such nonhalophytes as Cadaba farinosa, Dichrostachys glomerata and Adansonia digitata) in Senegal are also apparently living relicts of these much more widespread mangrove communities (Trochain and Dulau, 1942). As a consequence of the adverse sealevel and climatic changes since the last transgression, the mangrove vegetation has contracted significantly along the west coast of Africa in areal extent, with the remnants virtually confined to refugia such as lagoons, embayments and deltas. These relatively recent events must also be considered in relation to those of the geological past: a number of the proto-Atlantic mangrove taxa have migrated out of the region (e.g. Nypa, Mauritia and Kentia) while only selected members of the Indo-Pacific mangrove flora were able to enter and survive in the proto-Atlantic before the Tethys Sea (Mediterranean) became unavailable as a migration route. As mentioned earlier, at present there are only six indigenous species of mangroves on the Atlantic coast of Africa. This contrasts with the Atlantic coast of the Americas where there is an additional species of Avicennia (Table 4). This suggests that since the late Eocene (38 m.y. BP), when the protoAtlantic was being invaded by Indo-Pacific mangrove species, there has been complete and differential segregation between the Atlantic African and the East African/Indo-Pacific mangrove flora. However, since the mid-Miocene (20-14 m.y. BP), when the Tethys Sea dispersal route was no longer open (although it still contained extensive Avicennia communities with IndoPacific affinity - Bessedik, 1981; 1985) and direct dispersal between America and Africa was apparently no longer possible, there has been speciation only in the genus Avicennia in America, but apparently not in Rhizophora, Laguncularia or Conocarpus on either coast (Keay, 1953; Tomlinson, 1986; Jimenez, 1987). All of the indigenous species of the Atlantic coast of Africa are shared with the Atlantic and Pacific coasts of America. Obversely, there are no endemic mangroves (sensu stricto) on the Atlantic coast of Africa nor, for that matter, on the east coast of Africa where the mangrove flora consists simply of a subset of the Indo-Pacific species. In contrast, all other major mangrove regions have at least one or more endemic species (Table 4). The absence of any endemic species suggests that the African mangrove flora is a relict flora which, in turn, is supported by the fossil record which indicates that during the late Eocene to early Miocene, a much richer mangrove flora was present including, amongst others, Nypa, Mauritia and Pelliciera. To that must be added that the mangrove vegetation of the Atlantic coast of Africa is in retreat from relatively recent climatic and sealevel

changes. It also suggests that the conditions along this coastline were not conducive to trigger speciation in the taxa present and/or that continuous geneflow was maintained across the Atlantic - a point which is further discussed below. Biogeography The northern and southern limits of the main mangrove genera are shown in figure 3. Avicennia is recorded as far north as the le Tidra (1950'N) in Mauritania (Naurois and Roux, 1965), in an archipelago now included in Le Parc National du Banc d'Arguin, established on 24 June 1976. The genus Rhizophora is reported to extend to a little north of Saint Louis in the delta of the Tiallakh River (1611'N) in southern Mauritania (Naurois and Roux, 1965), although which of the three species is involved is not explicit. However, on the basis of the occurrence of only R. racemosa in southern Mauritania (Nicou, 1956; Adam, 1965), this record is almost certainly of R. racemosa. It should be noted, however, that the stands of R. racemosa reported by Adam (1965) from the le de Thiong, have been cleared by the local inhabitants in the last two decades (Gowthorpe and Lamarche, 1993). While the exact extent of Conocarpus erectus is not recorded, Adam (1965) reported its presence together with A. germinans and R. racemosa on the le de Thiong (1603'N) on the Mauritanian banks of the Senegal River. This probably comprises its northernmost occurrence. The northern limit of Laguncularia racemosa is uncertain although the records from the delta of the Senegal River (Nicou, 1956) apparently comprise its northern limit. The southern limit of extensive mangrove vegetation is found in the Angolan estuary of the Rio Longa (1018'S) although isolated occurrences of A. germinans occur slightly south of Lobito (1220'S) (Gossweiler and Mendona, 1939; see also partial English translation of this paper by AiryShaw, 1947). Gossweiler and Mendona (1939) also indicate that R. mangle does not occur south of Lobito where its niche appears to be occupied by L. racemosa, Scaevola plumieri and Suaeda. However, as these authors did not distinguish the various species of Rhizophora, precisely which species comprises the southernmost record remains unresolved. The most southerly record of Conocarpus in Angola is from Praia de Chiloanga (55'S) in the Cabinda Province of Angola (Gossweiler and Mendona, 1939). However, Pynaert (1933) records it from the northern banks of the Rio Zaire (610'S) which appears to be the most southerly record. The southern limit of L. racemosa lies between Lobito and Momedes (1510'S). Both the northern and southern limits of mangroves (fig. 2) coincide with the limits of arid regions (UNESCO, 1979) defined as: summer rainfall and winter drought, 12 months/year with <30mm rainfall, and a precipitation to potential evapotranspiration ratio (P/PEt) < 0.03. This suggests that, as found on other western coasts of continents (e.g. Australia - Saenger and

Moverley, 1985), mangrove distribution in western Africa is more limited by aridity rather than by temperature. In order to determine the potential dispersal distance of mangroves along the west African coast, the data on the distribution of mangrove and associated species on the offshore islands in the Gulf of Guinea (fig. 4) have been collated (Table 5) and regressed against the island area (fig. 5) and the distance from the mainland and nearest landward island (fig. 6). As expected, there is a more or less linear relationship with island area, suggesting that habitat availability is a significant factor. The analysis against distance suggests that the dispersal ability of mangroves under these equatorial conditions allows a dispersal distance of 350km from the mainland or 250km from another island. Given a dispersal longevity in Panama of 35 days for Laguncularia, 110 for Avicennia and around 300 for Rhizophora (Rabinowitz, 1978) and with current speeds in the Gulf of Guinea around 0.3m/sec, the dispersal distances obtained from this simplistic analysis seem realistic. In turn, however, these dispersal distances would preclude any continuous (and unassisted) geneflow across the Atlantic Ocean. Despite the extensive dispersal distances of these mangroves, plant distribution along the western shorelines of Africa is not uniform but shows a degree of patchiness (Table 6). For example, Rhizophora mangle is absent in Cte d'Ivoire, Ghana and Benin and is only very rare with only a small stand of R. mangle known from the lagoon at Owendo (Lebigre, 1983) in Gabon. In addition to their variable distribution along the coastline, variable distances have been recorded for mangrove penetration up river systems (Table 7), ranging from around 100km in Guinea-Bissau to less than 20km in Equatorial Guinea. These upriver distances probably reflect the seasonal differences in hydrology, with greater salinity intrusion during the dry season. Extent of Mangrove Vegetation Altogether, African mangrove vegetation covers 45,787km2 of which 31,111km2 is found along western African coastline, 3,315km2 in Madagascar and 11,361km2 in the remainder of eastern Africa. The best estimates of the areal extent of mangrove vegetation along the western coast of Africa are shown in Table 8, together with some demographic and economic indicators of the respective states. These data show that Nigeria and Gabon contain the largest mangrove areas and comprise over 50% of the total area for the Atlantic coastline. When the ratio of total mangrove area to coastline length in each country is considered (fig. 7), the mean breadth of mangrove vegetation ranges from 12.3km in Nigeria to a matter of metres in Mauritania. More significantly, however, the

data in figure 7 indicates that, based on the breadth of mangrove vegetation, five groups can be recognized as follows: 1. 2. 3. 4. 5. Nigeria Guinea Gambia, Guinea-Bissau, Cameroon, Gabon and Zaire, Senegal, Sierra Leone Mauritania, Liberia, Cte d'Ivoire, Ghana, Togo, Benin, Congo, Equatorial Guinea and Angola.

These groups, while artificial, clearly reflect the physiographic settings (described earlier) as well as the hydrological regimes of the regional mangrove communities. Consequently, these groups have been used for the regional descriptions below. Regional Descriptions of Mangrove Vegetation Group 1: NIGERIA This region is characterized by the high rainfall and humid conditions associated with the Niger delta which dominates the 853km long shoreline of Nigeria. The Niger delta has developed from the interaction of the sedimentladen Niger and coastal processes, thereby creating beach-ridge barrier islands, a freshwater floodplain, and between these structures an extensive brackish mangrove swamp (Keay, 1953; Savory, 1953; Wilcox, 1985; Adegbehin, 1993; Ajao, 1993; Isebor and Awosika, 1993). However, extensive lagoonal systems also occur around Lagos in the west and estuarine systems characterize the coast between the Niger delta and the Cameroon border (Amadi, 1990). In the lagoons and deltas, Rhizophora racemosa is the most abundant mangrove and Avicennia germinans, R. harrisonii and R. mangle are only sparsely present (Wilcox, 1985, CEC 1987; Isebor, 1993). In the estuaries, however, the mangrove communities may differ in species composition. Thus, in the Bonny estuary, Nypa is abundant near Port Harcourt but decreases in abundance towards the mouth where the mangroves rarely exceed 10-12m in height (Saenger, pers. obs.). Ukpong (1989; 1992) has described the mangrove communities of the Imo and Kwa Ibo Rivers in south-eastern Nigeria where there are approximately 196km2 and 50km2 of mangroves respectively. These communities are numerically dominated by A. germinans as shown in Table 9. Heights varied from 3-15m with A. germinans the tallest trees, particularly at the estuarine mouths. The abundance of Nypa fruticans in these plots indicates that, at least during the wet season, these areas are brackish to almost freshwater, while the dominance of A. germinans suggests that highly saline conditions occur during the dry season. Using ordination analysis, Ukpong (1992) concluded that six community types can be distinguished, although each of the types is co-dominated by A.

10

germinans and, at least, one of the Rhizophora spp. Ukpong (1989; 1992) also investigated the relationship between the vegetation and the soil characteristics at rooting depth. Basal area of the trees, tree density and cover were negatively correlated with soil chloride content. Group 2: GUINEA This region is also characterized by a high rainfall (4125 mm/y in Conakry) but the coastal configuration is less suited to the formation of extensive deltas. Nevertheless, most mangroves are associated with deltas and rias although frontal mangroves also occur, sometimes associated with cheniers and/or beach ridges (Schnell, 1952; Jacques-Felix and Chezeau, 1960; Bertrand, 1991). Saltflats or 'tannes' occur widely and are associated with chenier plains - but not generally with beach-ridge areas. Although Bertrand (1991) suggests that R. mangle is rare and only found at the mouths of some rivers e.g. Tabounsou and Rio Nunez, Jacques-Felix (1957) described it as widespread from the les Tristo in the north to the Sierra Leone border in the south. Structural analysis (Table 10) of the littoral communities in the delta of the Soumba River (945'N) based on SPOT image 30-330 taken on 19 April 1986 (fig. 8) showed that, even in a high rainfall deltaic area, the mangrove vegetation is dominated by short (<7m high) formations and that tall (>15m high) mangroves occupy an area of around 6.4% of the littoral vegetation.

3. GAMBIA, GUINEA-BISSAU, CAMEROON, GABON AND ZAIRE, The mangrove vegetation of this group is largely confined to estuaries with seasonally high rainfall and river flow which has led to considerable mudflat development. In these regions, the rivers constitute 'salinity refugias'. For example, the Gambia River, with its relatively high and continuous freshwater input, acts as a brackish water system throughout the year although the position of the salt wedge shows considerable up- and down-river movement (Blasco, 1983; Healey et al., 1988). In Zaire there are 226km2 of mangroves associated with the mouth of the Zaire River (Pynaert, 1933; Delevoy, 1945; Robyns, 1950; Sayer et al., 1992) but they do not extend north of the mouth except for one estuary (Darteville, 1950). Throughout these areas, R. racemosa is the pioneer species, colonizing recently deposited alluvial soils up to the level of daily tidal flooding. It may also form tall gallery forests (e.g. along the lower Gambia River or in the Wouri Estuary) reaching a height of 40m and 1m trunk diameters (Pynaert, 1933; Boy et al., 1975; Blasco, 1983; Din, 1991). However, according to Lebigre (1983), Rhizophora racemosa can be classified into three formations

11

in Gabon: a very tall form, such as at Pointe Pongw, varying between 2040m in height and resembling the gallery forests of Gambia while at the other extreme, there is a very small form, characteristic of the Rio Muni and Gabon estuary, often with a ground cover of Paspalum vaginatum or Eleocharis spp. R. harrisonii occurs with R. mangle at the boundary between Rhizophora and Avicennia mangrove stands. Giglioli and Thornton (1965), however, found that at Keneba, in lower Gambia, R. harrisonii colonizes the fibrous clay recently vacated by R. racemosa and that R. mangle grows in the higher elevation, sandier soils where it is often intermixed with the waterward edge of the Avicennia zone. R. harrisonii is intermediate in size (3-6m in Cameroon) between R. racemosa and the small form of R. mangle found in West Africa. In Gabon, Lebigre (1983) described this species as mostly of the order of 2m in height but there are some tall R. harrisoni bordering the creeks and in forests abutting the upland areas. Soils are acid-sulphate but little different from those of R. racemosa. R. mangle occurs in the drier and often more saline landward edge of the Rhizophora zone. Avicennia typically occurs on the higher elevation fibrous clay or sandier soils. A. germinans may vary in height from 7-20m in Gambia (Blasco, 1983), around 12m in Cameroon (Din, 1991) or up to ~30m with trunks around 1 m in diameter and virtually no understorey in Gabon (Lebigre, 1983). Laguncularia racemosa is uncommon but may form a shrub <3m high, generally co-occurring with low-growing Rhizophora or Avicennia. Other mangrove associates commonly present include Hibiscus tiliaceus, Conocarpus erectus, Carapa procera, Drepanocarpus lunatus, Dalbergia ecastaphyllum, Oxystigma mannii, Pandanus candelabrum, Phoenix reclinata, Pterocarpus and Acrostichum aureum (Pynaert, 1933; Delevoy, 1945; Boy et al., 1975; Blasco, 1983; Lebigre, 1983; CEC 1987; Din, 1991; Appolinaire, 1993). In Cameroon, Nypa has recently become distributed throughout the mangroves of the Wouri estuary (Saenger, pers. obs.) and Din (1991) records its presence at Pointe Olga. Saltflats or 'tannes' occur throughout this region wherever there is a marked dry season. Giglioni and Thornton (1965) suggested that where evapotranspiration exceeds 1500mm per year, barren saltflats replace herbaceous halophytes. Although most saltflats are probably of human origin, particularly in the higher rainfall areas, Anthony (1989) has suggested that their extensive occurrence, notably north of the Guinea coast, must reflect, at least partly, a combination of a sealevel pulse in the last 1000 years (from a highstand of 1-1.5m to the present level) and climatic deterioration (i.e. increasing aridity). The saltflats generally consist of clayey soils, which may be fairly acid, or more rarely of sandy soils (Lebigre, 1983). These areas generally have

12

mangroves (often <1m high Avicennia, Laguncularia or Conocarpus) around their margins with patches of Sesuvium portulacastrum and sedges or grasses (such as Paspalum vaginatum). At their landward margins palms, Hibiscus tiliaceus or Pandanus candelabrum are commonly present. Both in Cameroon and in Gabon, a freshwater swamp forest appears to be associated with the mangroves (Boy et al., 1975; Lebigre, 1983). This formation is generally saline but often flooded by rains due to poor drainage. Rhizophora, Pandanus and Acrostichum aureum are present, but also the following rainforest species: Chrysobalanus ellipticus (Rosaceae), Manilkara lacera (Sapotaceae) and Syzigium guineense (Myrtaceae). Group 4: SENEGAL AND SIERRA LEONE This mangrove group is transitional between the seasonally wet group (Group 3) and the dry-arid group (Group 5) discussed below. In these regions, the mangroves are predominantly deltaic but the more fluvial elements (such as the freshwater swamp forests) are absent. Mangroves in Senegal are concentrated in two main areas, each typified by a particular river - the Casamance and the Saloum (Marius, 1984). Both these estuaries are 'inverse estuaries'. During the dry season, the salinity of the waters around the mangroves can be 2-3 times greater than that found in neighbouring seawater (Marius, 1981; 1984; Marius and Lucas, 1982). In Sierra Leone, approximately 80% of the total mangrove vegetation occurs in the estuaries of the Scarcies, Sierra Leone and Sherbo Rivers (Johnson and Johnson, 1993). Tomlinson (1957) has described the mangroves of northern Sierra Leone, close to the border with Guinea while Gledhill (1963) has described those of Aberdeen Creek. At Aberdeen Creek, R. racemosa is rare (Gledhill, 1963) but R. harrisonii may be up to 5m, R. mangle up to 2m, Laguncularia to 4m, Avicennia to 5m and Conocarpus 2-3m. As elsewhere in West Africa, R. racemosa generally occurs on newly deposited sediments fringing creeks and may be tall while the other species are all low (Thompson, 1945; Tomlinson, 1957; Jordan, 1963; 1964). The occurrence of R. racemosa on new deposits may be due to a high nutrient demand of this species or to the difficulty of the establishment of its seedlings in the fibrous soil. As R. racemosa matures, the soils invariably become highly fibrous while under the other species it is more variable (Hesse, 1961). Where shrubby growth of Avicennia occurs, it may be growing on fibrous soils after the initial colonizing R. racemosa has died or been removed. Thompson (1945) has described mangroves in Sierra Leone as consisting exclusively of Rhizophora wherever marked freshwater seepage occurs. In contrast, Avicennia may co-occur with Rhizophora where freshwater seepage is absent.

13

It should also be noted that Adam (1958) reports the presence of the mangrove creeper Brachypterys ovata (Family Malpighiaceae) from Sierra Leone and Guinea (near Dubrka and Conakry). The restricted distribution of this species in West Africa, together with its widespread occurrence on the Atlantic coast of tropical America, suggests that this species may be a relatively recent introduction which requires further investigation. Limited salt-marshes occur in Senegal and Sierra Leone with the following species: Sesuvium portulacastrum, Paspalum vaginatum, Eleocharis caribaea, E. mutata, Sporobolus robustus, Philoxerus vermicularis and Scirpus maritimus (Marius, 1979; Tomlinson, 1957). Group 5. MAURITANIA, LIBERIA, COTE D'IVOIRE, GHANA, TOGO, BENIN, EQUATORIAL GUINEA, CONGO AND ANGOLA This group represents the most dry-arid mangrove vegetation on the Atlantic coast of Africa, where the mangroves are confined to narrow fringes partially lining estuaries and coastal lagoons. Liberia clearly does not belong to this group on rainfall data alone but the virtual confinement of mangroves to deltas and the overall restricted area of mangroves in Liberia, has reduced the mean breadth of mangrove vegetation to that of other members of this group. In Mauritania, remnant stands of mangroves have been described consisting only of Avicennia germinans (Naurois and Roux, 1965; Gowthorpe and Lamarche 1993). Liberia does not have extensive wetlands along its coasts and most mangroves occur at river mouths with some lesser development of lagoonal communities (Kunkel, 1966). Thus, Adam (1970a) has described the lagoonal mangroves around Cape Palmas in south-eastern Liberia. These communities attain a height of around 3m and are dominated by Conocarpus erectus with only rare specimens of Avicennia germinans and Rhizophora racemosa. Thickets of the fern Acrostichum aureum are also common. Near Buchanan on the central Liberian coast, estuarine mangroves occur (Adam 1970b). Although these may have once been dominated by Rhizophora racemosa, this species is locally absent due to felling. The remaining vegetation consists of stunted Rhizophora harrisonii, Avicennia germinans and Conocarpus erectus with thickets of Acrostichum aureum. Laguncularia racemosa was not observed. Kunkel (1966) has described the mangrove-associated flora from Liberia and has interpreted its relationship in strictly successional terms. Whether the Liberian successional sequences are really applicable elsewhere is doubtful, particularly where the rate of sand movement and erosional mudflat deposition is lower than in Liberia.

14

In Cte d'Ivoire, the climatic variation along the coast is very slight and physiognomic/species differences must be non-climatic e.g. hydrological, geomorphological etc. Egnankou (1985) has subdivided the mangroves of Cte d'Ivoire into two groups: (1) Those from Assinie to Fresco (including lagunes Aby, Ebri, Grand Lahou, and Potou as well as the M, Comoe, Bandama, Tagba Rivers all of which empty into the coastal lagoon system) These lagoonal mangroves are relatively depauperate and only around Abata to Mossou and around Grand Bassam, do they attain heights of around 20m. Only R. racemosa, Avicennia and Conocarpus occur with the best stands on River M. R. racemosa colonizes the front of the water. (2) The second group are the deltaic rivermouth systems associated with the Sassandra, Cavally, Niro, San Pdro, Brim, Dodo, Nichia and Tabou on the coast between Fresco and Bliron. These are dominated by A. germinans and R. racemosa. Paradis (1989) has described the mangrove communities of the lagoonal system around Fresco in detail, particularly the lagune Nyi. He described four zones as including (1) an Avicennia zone 15-18m high; (2) a Rhizophora racemosa zone 8-10m high; (3) a shrub zone of Dalbergia 4m high (mixed with H. tiliaceus, Drepanocarpus and Acrostichum); and (4) a Paspalum vaginatum grassland 0.8m high. Hdin (1933) has described the nonmangrove lagoonal vegetation around Grand Bassam. Virtually all of the mangroves of the lagoonal system of Cte d'Ivoire have been hydrologically disturbed, particularly by the construction of the canal de Vridi, commenced in 1938 and completed in 1950 (Paradis, 1989). It is 2.7km long and provides a permanent opening to the lagoonal system where it did not previously exist - while the old mouth at Grand Bassam is now generally sanded up (Durand and Skubich, 1982; Durand and Chantraine, 1982). The climate of Ghana is relatively dry and much of the coastal areas are dominated by savannah with occasional boababs (Adansonia digitata). Mangroves are confined to the lower reaches and delta of the Volta River and to the numerous coastal lagoons such as Keta, Korle and Kpeshie Lagoons (Kwei, 1977; CEC 1987; Amlalo, 1990). Boughey (1957) has suggested that a clear distinction in terms of species composition could be made in Ghana with species of Rhizophora confined to open lagoons. Closed lagoons, characterized by elevated salinity during the dry season, support populations of the more salt-tolerant species including Avicennia, Conocarpus, Laguncularia and Acrostichum. This explanation is similar to that offered by Thompson (1945) for Sierra Leone, where Rhizophora and Avicennia occurred together wherever salinity became elevated due to the absence of freshwater seepage. While such a distinction in the lagoonal setting is generally helpful, there are exceptions. For example, Korle Lagoon near Accra, has an artificially

15

maintained connection to the sea but the mangroves are dominated by Avicennia germinans to the virtual exclusion of Rhizophora. All mangroves in Benin are lagoonal as there are no active deltas and the open shoreline is sandy and of relatively high energy (Paradis, 1976; Baglo, 1989). Rhizophora mangle is absent in Benin which Blasco (1985) has suggested might be due to the irregular and hyposaline hydrological regime within the lagoonal system of Benin. Laguncularia and R. harrisonii are rare (Paradis and Adjanohoun, 1974; Blasco, 1985; Baglo, 1989). Mangroves have virtually disappeared from Lac Nokou due to the Cotonou barrage, and the general regional aridity. In Lac Ahm there are only ~50ha left. Extending for 60km from Togbin to Grand Popo in the west, the coastal lagoon system contains the most important mangrove stands (Paradis, 1981). This lagoonal system is connected to the sea by a small opening which closes naturally, causing the Mono River (mean flow 500-1000m3/sec) to flow inland to Lac Ahm during the wet season. When flooding becomes extreme, local fishermen open a new connection. The mangroves of this lagoonal system consists of Rhizophora and Avicennia 5-6m high with occasional emergents up to 15m. At the back, Laguncularia, Dalbergia, Drepanocarpus, Paspalum and Acrostichum aureum. Only at Azizahoue can trees still be found up to 25m in the coastal lagoon (CEC, 1987). At Adounko (3km east of Togbin) there is a small stand of R. harrisonii around 20m high while around Gbetou and le d'Ahoundji in Lac Ahm, there are mixed stands of Avicennia and Rhizophora between 8-13m high (Paradis, 1981; Blasco, 1985). There is one private mangrove plantation in the lagoon at Ahouandji, planted by a local fisherman to enhance his regular catch. Seedlings of Rhizophora were planted and these have been augmented by natural seedlings of Avicennia (Blasco, 1985). In the lagoonal system of Togo, mangroves occur around the mouth of the Mono River and in its western tributaries, particularly the Vensi and Gbaga Rivers (Paradis, 1981; Akpagana, 1993). Formerly, they appear to have occurred further east as Grewe (1941) records mangroves in Kitta Lagoon and on the eastern shores of Lake Togo. Mangroves are confined to estuaries in Equatorial Guinea with the following major areas: Ntem R. 18km2; Mbia R. 1.5km2; Ekuko R. 10km2; Mbini 53km2 Muni 65km2 (Sayer et al., 1992). Similarly, mangroves of Congo occur within estuaries and lagoons of the Atlantic coast (Camara, 1993). Mangrove communities are extensively developed in the Conkouati and Malonda lagoons but also occur in the estuaries of Noumbi, Lom and Kouilou. According to Lebrun (1954), the mangroves are dominated by Rhizophora racemosa with Avicennia, Conocarpus and Laguncularia occurring respectively in order of frequency.

16

The most extensive stands in Angola are found in the Rio Lubinda (55'S), Cabinda, and in the Rio Zaire (610'S) where they appear to extend along the southern open coast. Other extensive mangrove communities occur at the mouths of the Chiluango, Loge and Cuanza Rivers with smaller outliers appearing southwards to Benguela. R. racemosa and R. mangle can reach heights of up to 30m in the north of Angola (Huntley 1974), while south of Cuanza (915'S), Rhizophora is very short with heights of <1m (Gossweiler and Mendona, 1939). Avicennia attains a height of 3-5m in the delta of the Rio Zaire but forms stunted formations further south (Gossweiler and Mendona, 1939). North of Cuanza, the landward margins of the mangroves are occupied by a littoral rainforest which appears to be similar to the freshwater swamp forests ('fort inondee') described by Lebigre (1983) from Gabon with the following species in Angola (Gossweiler and Mendona, 1939): Chrysobalanus ellipticus, Drepanocarpus lunatus, Dalbergia ecastaphyllum, Pandanus, Annona palustris, Hibiscus tiliaceus, Erythroxylum emarginatum, Flagellaria indicum, Raphia spp. and Phoenix canariensis. Herbaceous halophyte species recorded from Angola (Gossweiler and Mendona, 1939) include: Arthrocnemum fruticosum, Arthrocnemum indicum, Arthrocnemum macrostachyum, Cyperus maritimus, Paspalum vaginatum, Sesuvium portulacastrum, Sporobolus robustus, Sporobolus virginicus and Suaeda fruticosa. Phenology There are no systematic data on phenological events in mangroves from western Africa. The limited data on flowering of the various species at different localities (Table 11) have been compiled from incidental records in taxonomic or ecological accounts. These data indicate, however, that there are no clear trends of shifting flowering periods between north and south. When the flowering records are arranged according to month for all locations combined (Table 12), the records indicate that slightly more than 60% of flowering occurs in the dry season, approximately from November to April north of the equator and May to October to the south. This finding is supported by Din (1991), who reported that flowering in Cameroon occurs predominantly during the dry season, i.e. mid-November to mid-March. Amobi (1974) also reports dry season flowering in the mangrove associates Dalbergia ecastaphyllum, Drepanocarpus lunatus and Ormocarpum verrucosum. As with flowering, there are no systematic data on fruiting although Jackson (1964) provided data for seed-fall for one site in Nigeria. These data (Table 13) indicate that at Ikorodu in Western Nigeria, where the pronounced dry season extends from January to April, most species shed their propagules towards the end of the wet season and only R. racemosa and H. tiliaceus do so late in the dry season. However, the data are too scanty to indicate

17

whether or not regional trends of shifting time patterns occur. However, when all of the available fruiting data are combined irrespective of locality (Table 14), mature fruit appear to be present predominantly during the wet period, i.e. May to October to the north of the equator, with slightly more than 60% of the records occurring in those months. This suggests that the time sequence described by Jackson (1964) occurs throughout the region. It appears from the above that the principal mangrove species and their associates flower during the dry period of the year and fruit largely during the wet season. Given that all of the listed species have water-dispersed propagules and that they are either wind- or insect-pollinated, this phenological pattern is not surprising. Physiology There are virtually no physiological data available from West Africa on indigenous mangrove species and any associated flora. The only exception appears to be the osmotic data obtained by Walter and Steiner (1936) on their brief visit to Cameroon while returning from East Africa. As a consequence, any physiological insights into West African mangrove systems have been derived by extrapolating North American data of shared mangrove species. While such extrapolations may be useful, they are based on the underlying assumption that the West African species do not differ physiologically from their North American analogues. While this assumption is based on the absence of any noticeable morphological difference, it has yet to be established that no physiological, biochemical or genetic differences exists. Comparative investigations into the biochemistry and genetics of Avicennia germinans from the Atlantic coasts of Africa and America are presently underway at ICIV, Toulouse in co-operation with the University of California (Berkeley). Given that the mangrove populations on the two sides of the Atlantic have been separated for at least 10 million years with little likelihood of any significant geneflow between them, is it realistic to assume physiological, biochemical or genetic equivalence? If anything, a re-examination of the original data of Walter and Steiner (1936) suggests that there are some physiological differences in the mangrove plants of the two sides of the Atlantic. Walter and Steiner (1936) investigated the osmotic concentration of mangroves using a (then) newly developed refractometric technique. They simultaneously measured the chloride osmotic concentration in their samples. These measurements were carried out on a range of mangroves and associated species in Tanzania (East Africa) and Cameroon (West Africa) and they tabulated results obtained by the same technique from a range of mangrove species in Florida and the Caribbean.

18

These data have been summarized (fig. 9) and suggest that in the West African species, the total osmotic concentration directly varies with, and is accounted for, by the osmotic concentration due to chlorides. In contrast, the same species from either East Africa (Terminalia catappa, Pandanus candelabrum, Phoenix reclinata, Acrostichum speciosum and Hibiscus tiliaceus) or North America (Rhizophora mangle, Avicennia germinans, Conocarpus erectus and Laguncularia racemosa) showed that with increasing total osmotic concentration, no linear relationship with chloride concentration exist and that the chloride concentration could not explain the total osmotic concentration. Given the current understanding of osmo-compensation in mangroves (Wyn Jones and Storey, 1981; Robinson and Jones, 1986), increasing osmotic concentration in the expressed cell sap is not accompanied by increased chloride but, rather, by a series of organic compounds such as glycinebetaine, choline-O-sulphate, choline-O-phosphate, proline and sorbitol. These osmo-compensators serve to maintain the osmotic concentration of the cell sap without increasing chloride concentrations which could adversely affect enzymes in the cytoplasm. What the West African mangrove data suggest is that the ability to osmo-compensate is less well-developed in the West African mangrove species in comparison with their analogues elsewhere. Because of the widespread occurrence of osmocompensators in other mangrove regions, it appears that any lesser ability to be able to osmo-compensate is due to a secondary loss perhaps due to genedrift in remnant populations of West Africa. The ecological implications of such a reduced ability to osmo-compensate in West African mangroves includes the pronounced confinement of mangroves to salinity refugia, and the susceptibility of mangrove communities to periods of aridity or increased saltwater incursion (such as might be expected with rapid sealevel rise). Mangrove Zonation In western Africa, mangrove zonation can only be recognized on a broadly generalized scale as clear zonation is absent at many sites (Din, 1991). Numerous irregularities in the vegetation exist because different species tend to establish on different microtopographic features and soil types. As a result, there appears to be a dynamic equilibrium with the vegetation constantly adjusting to the ever-changing estuarine landscape (Ukpong, 1992). Nevertheless, R. racemosa is generally the predominant species on recently deposited, unconsolidated alluvium (Rosevear, 1947; Savory, 1953; JacquesFelix, 1957; Jordan, 1964; Villiers, 1973; Marius, 1979; Lebigre, 1983; Egnankou, 1985; Din, 1991), often forming a monospecific zone. This monospecific zone is generally followed by a mixed Rhizophora zone, in which R. racemosa, R. harrisonii and R. mangle may co-occur, where the sediments are more consolidated but still inundated by the daily tides.

19

Above this level, where tidal inundation is reduced, an Avicennia zone generally occurs which may be monospecific, mixed with Laguncularia or Conocarpus, or interspersed with herbaceous halophytes with Paspalum vaginatum and/or Sesuvium portulacastrum. The more open Avicennia zone may gradually be replaced by Paspalum vaginatum or Sesuvium portulacastrum which, with further soil build-up, may result in vegetationfree saltflats (or 'tannes') in those areas inundated only by extreme spring tides each month. This characteristic seaward to landward sequence of species is illustrated (fig. 10) by data from Gabon (Villiers, 1973). In contrast to zonation elsewhere (Snedaker, 1982), this sequence of species may not reflect intertidal gradients of waterlogging or salinity but may be due to varying soil preferences of the different species. For example, the prop roots of R. racemosa form a flying buttress structured base and below the surface of the soil, the roots immediately divide into innumerable hair-fine rootlets - supporting the tree by a thickly intertwined mat of roots (Attims and Cremers, 1967). This root mat is the source of Rhizophora peat deposits which, as Rosevear (1947) noted, may be so dense that it can be cut out in blocks, dried, and burned as peat fuel. With further growth of R. racemosa, the thick root mat gradually raises the soil surface which reduces the tidal flushing and may ultimately lead to the death of this species. Rosevear (1947) suggested that R. racemosa can only reach its optimal size where it starts growing on newly deposited soft mud. It may be that only one crop of the tall R. racemosa can grow at any one site before the root mat and elevation of the soil surface make the site unsuitable for a subsequent crop of this species (Rosevear, 1947; Jordan, 1964). Generally scrubby growth of Rhizophora follows tall R. racemosa. R. harrisonii is often dominant in the middle areas of the Rhizophora zone. It does not compete with R. racemosa as a pioneer, apparently preferring the wetter, middle parts to the firmer and more peaty soil towards the inner limit (Savory, 1953). It rarely forms pure stands (Villiers, 1973). Where it occurs, R. mangle is generally found on the drier, inner limit of the Rhizophora zone in western Africa although Lebigre (1989) records it only from one river mouth in Gabon. While it forms extensive pioneer forests of large trees in America, so far there are no confirmed reports of large trees of R. mangle in the mangroves of western Africa and Savory (1953) reported that, in Nigeria, this species grew only as a scrub up to 4 m high. Savory (1953) first suggested that this difference in the distribution of the three species of Rhizophora lies in their different salt tolerances. Indeed, Chapman (1944) had shown that on the Jamaican shoreline, R. mangle had a high salt tolerance while R. racemosa has a lower salt tolerance. Thus, in Africa where seasonal freshwater inundation is common, R. racemosa is the dominant species and R. mangle is confined to higher land - which is less subject to flooding during the wet season with significantly higher salt concentrations during the dry seasons. On the other hand, Jordan (1963) has ascribed the distribution of this species to its inability to tolerate shade

20

and, therefore, it is unable to compete with R. racemosa. It co-occurs with Avicennia (which has a more open canopy) and also grows in those areas (e.g. more landward and often sandier fringes) where R. racemosa does not occur. Tomlinson (1957) noted that "the factors which control the distribution of the different species of vegetation are incompletely understood. It is usual to find R. racemosa fringing creeks in almost pure stands. Avicennia nitida is often found on slightly higher land away from the creek but seldom occurs as pure stands. The indications are that while R. racemosa makes good growth only on recently deposited heavy soils, Avicennia is particularly salttolerant and grows well on rather sandy soils in which there is marked accumulation of salt during the dry season, at the end of which salt crystals are commonly found on its leaves. The grasslands are still further removed from the creeks. The soils have a high proportion of sand, commonly about 70% at the surface, and are dry during the dry season." In Nigeria, A. germinans cannot be considered as a frontal species but Rosevear (1947) described one such formation on Soden Island in the estuary of the Rio del Rey. Here Avicennia grows on sand, reaching heights of slightly over 30m with a clear bole length of 22m and girths up to 1.5m. On newly formed banks at the mouths of creeks opening into the sea numerous associations of Laguncularia bushes occurred, apparently as first colonists. Behind these, Avicennia was found. Rosevear (1947) concluded that "it seems possible that these two species may establish themselves on deposits which are predominantly sand from the sea rather than silt from the river, and that behind these barriers the normal slimy mangrove soil collects on which Rhizophora finds secure and favorable anchorage." From his studies in Gabon, Lebigre (1983) also concluded that where rapid colonization was taking place, the pioneer species are Laguncularia and Avicennia but these are rapidly submerged or eliminated by Rhizophora. Similarly, Paradis (1979) has described Laguncularia and Dalbergia as 'opportunists' which characterize secondary successional stages in mangroves. As both are shade-intolerant and have abundant, widely dispersed seeds, these two species are able to rapidly colonize any gaps formed in the mangroves due to human disturbance. Particularly in the case of Laguncularia, this helps to explain the various accounts of its distribution in relation to the other species: for example, Rosevear (1947), Lebigre (1983) and Din (1991) describe Laguncularia as a pioneer species, colonizing newly deposited mudflats in the very frontal zone. Others describe it as occurring at the landward boundary of the mangroves where soils are better drained and of lower salinity; Laguncularia seems to be able to colonize virtually any unoccupied area in sheltered intertidal zones, becoming outcompeted or excluded by other species with denser canopies. Part of this wide ecological ability may be attributable to the abundant peg-root development (Jenik, 1970) that occurs in Laguncularia when growing on newly deposited muds.

21

Lebigre (1983) has additionally described the changes in zonation that were noted with increasing distance up-river in Gabon. Thus, near the mouth of the rivers, a zone of tall R. racemosa was noted, followed by a landward zone of Avicennia. In the mid-zone of the rivers, R. harrisonii predominated over R. racemosa, with well-developed Avicennia zones on their landward sides. Using air-photo interpretation, it was also noted that abundant salt-flats or 'tannes' characterized the mid-zones of rivers. Upper zones of rivers were characterized by the predominance of R. racemosa along the river banks, the absence of a landward Avicennia zone, and the direct transition from R. racemosa to a freshwater swamp forest (Lebigre, 1983). A similar transition, directly from R. racemosa to a freshwater swamp forest, has also been reported from Cameroon (Boy et al., 1975). As discussed earlier, Kunkel (1966) has interpreted zonation in Liberian mangrove and associated vegetation in a strictly successional sense. This causative explanation has not been as prominently invoked in West Africa as it has been elsewhere and, as suggested by Snedaker (1982), is of doubtful general application. Structure and Productivity Only limited structural and productivity data have been collected from West African mangrove vegetation and one productivity estimate is available for Paspalum vaginatum saline grassland (Edwards and Ekundayo, 1982). The only biomass data for West African mangrove vegetation have been collected from the Saloum River (1440'S) in Senegal (UNESCO, 1983; Diop and B, 1993) using an allometric technique. Above-ground biomass of tall formations (mean height of 7.60m) was found to be 79t/ha while for medium formations (mean height of 5.64m), a biomass of 55t/ha was found (UNESCO, 1983). Diop and B (1993) provide some structural data for the Saloum River site for which biomass estimates have been determined, together with slightly different estimates of biomass: they report an aboveground biomass of 60t/ha comprised of 52t/ha wood, 6.6t/ha leaves and 1.4t/ha buds together with a net primary productivity of 2,145kg/ha/y. Wood volume was estimated at 62m3/ha (Diop and B, 1993). These aboveground estimates of biomass conform with the worldwide trends (Saenger and Snedaker, 1993) of biomass in mangrove vegetation. Blasco (1983) provides the following field estimates for biomass (wood volume) and productivity of Gambian mangroves: estuarine mangroves (<10m high) were estimated to have a standing stock of 50m3/ha, with a primary productivity of 8t/ha/y; fluvial mangroves (>20m high) were estimated to have a standing stock of 150m3/ha, and a primary productivity 15t/ha/y. While the wood volume estimates are similar to those measured in Senegal, the estimates of primary productivity are markedly higher. This is probably due to the greater height of the Gambian mangrove vegetation. For instance, the joint Gambian-German Forestry Project (quoted by Blasco, 1983) found the following structural distribution of mangrove vegetation in the Gambia catchment:

22

Community Mangrove type I Mangrove type II

Height (m) >7 <7

Dominant Rhizophora Avicennia

Area (ha) 15,312 51,067

% 23.1 76.9

The percentage (23.1%) of mangroves >7m high does not differ greatly from the remotely sensed data from Guinea (Table 10), where mangroves of greater than 7m in height comprised 23.8% of the total mangrove area. However, with approximately a quarter of the mangroves in Gambia exceeding 7m in height and reaching up to 20-25m, higher primary productivity in comparison to Senegal is to be expected. In terms of wood volume, the only other estimates are those of Pynaert (1933) for Zaire as follows: tall Rhizophora 500m3/ha; other mangrove types 100-400m3/ha. These estimates appear to be very high. On the other hand, in terms of the annual production of wood, Heske (1937) showed a mean annual increment (MAI) in mangroves on the Cameroon coast of 0.083m3/ha/y compared with an MAI of 1.98m3/ha/y for upland rainforest. These MAI estimates are very low when compared with MAI measurements elsewhere (e.g. 2-8m3/ha/y for Avicennia in Bangladesh - Saenger and Siddiqi, 1993) and, with the high estimates for wood volume, suggest an improbably ancient age for these mangrove communities. Other structural data are presented in Tables 9 and 15 for mangrove communities dominated by Avicennia and Rhizophora respectively. Both communities show a stem density of 384-492 stems/ha. Mean stem diameters (Table 15) for the Rhizophora dominated community varied from 8 to 31 cm while no data are available for the Avicennia dominated community (Table 9). However, Paradis (1989) has given the diameter distribution for Avicennia from 15-18m high stands at Fresco, Cte d'Ivoire, as follows:
class (cm) No. of trees 0-15 23 16-30 10 31-50 7 51-70 12 71-90 7 91-110 7 >110 1

While this undoubtedly represents an old, well-developed stand of Avicennia, there is the suggestion of a bimodal diameter distribution with a few very large old trees and more recent recruits in the <30cm diameter class. Mangrove Fauna and Trophic/Functional Aspects The mangrove vegetation of western Africa is intimately linked to the offshore coastal ecosystem (John and Lawson, 1990). Regularly influenced and disturbed by seasonal freshwater and diurnal tidal flooding, it exhibits features of an immature ecosystem, namely low species diversity and high productivity. The excess organic production is exploited by many marine species especially fishes and crustaceans that enter the mangrove environment as juveniles and return to the sea as adults for reproductive purposes.

23

The bacterially-mediated process by which this organic production becomes available for fish and crustaceans, has only been studied in the Gambia River. Healey et al. (1988) sampled the Gambia River during four different hydrologic seasons to determine selected microbial, nutrient, and physical parameters. The ranges of total bacterioplankton densities were similar to those of other tropical and temperate environments and the numbers of free bacteria remained similar over all seasons. However, the numbers of attached bacteria were greater during periods of high stream flows when suspended solids concentrations were higher. Consequently, heterotrophy overshadowed autotrophy in the river and bacterial abundance, distribution, and glucose uptake activity in this tropical floodplain river were greatly influenced by the annual flood. In turn, these data suggest that much of the organic material derived from the mangroves is broken down within the river and then flushed into nearshore waters with the annual flooding (Healey et al., 1988). Another linkage to the offshore coastal ecosystem is via the coastal waders which use many of the mangrove areas and coastal lagoons (Ntiamoa-Baidu, 1991). The significance of this linkage has only recently been recognized in West Africa with the identification of a series of potentially suitable sites for RAMSAR nomination in Cte d'Ivoire, Ghana and Benin (Schwarz, 1991). Major Threats to Mangrove Vegetation Prior to the colonial era, the mangrove vegetation of western Africa had little commercial significance although it had been traditionally exploited at the subsistence level. At the turn of the century, interest in the extraction of tanbark from mangroves was stimulated by analyses which showed a generally high (10-30% per dry weight) tannin content (Drabble and Nierenstein, 1907; Drabble, 1908; Pynaert, 1933; Grewe, 1941) while in Guinea, the combined exploitation for tan-bark, telegraph poles and railway sleepers was advocated (Baillaud, 1904). The first systematic exploitation of mangroves in western Africa, however, commenced at Manoka, Cameroon in 1919 when the 'Socit Nationale du Cameroun' was granted logging concessions and built a timbermill (Hdin, 1928). Considerable quantities of Rhizophora timber were extracted, initially as railway sleepers but, subsequently, for barrel staves which were ultimately used to export palm oil. The data shows a sizeable export of Rhizophora timber (Table 16) but the proportion that this represents of the total timber extracted is no longer ascertainable. Combined wood and tan-bark extraction also occurred in Cameroon, Gabon and Zaire but, in comparison with upland forest exploitation, it was relatively insignificant. Moreover, due to their somewhat lower tannin content, the mangroves of western Africa were spared the large-scale destruction that followed the intense tan-bark extraction in Mocambique, Tanzania and western Madagascar between 1905 to 1930 (Grewe, 1941).

24

As a result, the mangroves of western Africa represent a remnant system struggling against geological and climatic adversity but which has been spared the colonial over-exploitation that occurred elsewhere. Nevertheless, the demands of modern, post-colonial Africa, with its burgeoning population increasingly in urban centres, is making novel and threatening demands on these mangrove systems. These demands include extraction of products such as wood, often at levels which exceed sustainable levels; the use of mangrove lands for other productive (e.g. salt and/or rice production; aquaculture) or non-productive (e.g. foreshore and urban developments) activities; the accidental or deliberate discharge of wastes; and the alteration of hydrological regimes either through the upstream diversion of water by dam construction, or the altered influx of tidal waters through coastal erosion works or salt barrages. Each of these threats are briefly reviewed below: Domestic Firewood The extraction of wood for domestic purposes as well as for salt production and fish smoking is common throughout West Africa (Adam, 1958; Kinako, 1977; Egnankou, 1985; Paradis, 1989; Bertrand, 1991). For example, Adam (1958) reported that 25-30m high R. racemosa are heavily exploited in Gambia and that much of the wood and bark is marketed through Dakar. Similarly, Paradis (1989) described from air-photo interpretation that between 1962-76, Rhizophora racemosa areas around Fresco in Cte d'Ivoire, have been reduced by about 80% due to firewood collecting and for tan-bark which is used to prolong the life of wood used for pirogues. As a consequence, extensive prairies of salt-tolerant grass Paspalum vaginatum have increased dramatically. In Guinea, annual estimates of mangrove wood consumption (Diallo, 1993) are: rural firewood 152,000t/y; urban firewood 54,000t/y; fish smoking 58,000t/y; and salt extraction 93,000t/y. In Benin, domestic firewood usage is also high while additional demands result from "Acadja fishing", which is common in the lagoons and which requires additional mangrove wood for the stakes. The World Bank has estimated firewood consumption of Cotonou and Porto Novo as follows: 1984: 300,000m3/y actual; 2000: 650,000m3/y predicted; 2020: 1,400,000m3/y predicted. Such demands as in Guinea or Benin cannot be met sustainably without large-scale mangrove afforestation. In the Niger Delta, R. racemosa was heavily exploited for pitprops (e.g. at the Enugu coal mine, where 20,000m3 was used - Ajao, 1993), for poles, for firewood and by the local fishermen for tanning their fishnets (Rosevear, 1947). The wood of Avicennia was rarely used although there is some localized use of the leaves for preparation of salt. However, present day exploitation is largely confined to domestic firewood and estimates of the standing volume of 283.2 x 106m3 are considered to be capable of providing a sustainable timber yield of 0.6 to 1.6 x106 tonnes/y (Kinako, 1977).

25

Salt Production Salt extraction from West African mangrove areas is widespread (Paradis, 1979; Vanden Berghen, 1984; Blasco, 1985; Bertrand, 1991). For example, in Benin, salt production is a traditional industry where the short dry season is not sufficient for solar extraction. To produce salt, the mangrove vegetation is removed in selected areas and the underlying soils are dug up to a depth of around 10cm to facilitate evaporation. After a few days, the upper crusts are raked and scraped and, finally, collected into heaps for processing. These soils are leached several times with the same water and the saline solution is then boiled to obtain crystallized salt (Blasco, 1985). The firewood requirement is met by the neighbouring mangroves. Hachimou (1993) has estimated that 1m3 of mangrove wood is required to produce 100kg of salt. The environmental consequences of salt production in Benin, as in much of West Africa, include the formation of barren depressions in which the hydrological regime is modified; destruction of mangroves for firewood; compaction of the surface soil by repeated raking and trampling; and the creation of ecological conditions unsuitable for the recolonization by mangroves (Blasco 1985). Similarly, Paradis (1980) has suggested that where there are long dry seasons, bare salty areas are present and which may be extended (e.g. in Senegal and Gambia) by salt extraction. Even in more humid climates such as in Sierre Leone and Guinea, salt extraction has caused the formation of extensive denuded areas (Paradis, 1980; Bertrand, 1991). Rice Cultivation Using mangrove lands for the cultivation of rice is widespread in Gambia, Guinea, Guinea-Bissau, Senegal and Sierra Leone (Bertrand, 1991; Diallo, 1993). According to Simao (1993), the mangrove area in Guinea-Bissau has been halved due to ricefields and that the annual loss due to rice-growing was around 3,800ha/y between 1976 and 1990. In less humid regions, rice cultivation is more restricted (e.g. in Cte d'Ivoire, rice cultivation occurs around Sassandra - Egnankou, 1985) but still encroaches on the more limited deltaic mangrove vegetation. As many of the soils associated with mangroves have a high fibrous content, they often yield very low pH after drying out (Dent, 1947; Hart, et al., 1963; Thornton and Giglioli, 1965). Extensive acid-sulphate soils (Table 17) have caused difficulties with cultivation of dryland rice varieties in most regions of West Africa (Dost, 1988). Aquaculture Aquaculture is not widespread in West Africa at present but it can be expected to increase in the near future. In those localities where aquaculture

26

is already practised in mangrove areas, problems have been noted. For example, in Nigeria, aquaculture is being actively encouraged in mangrove areas (Wokoma and Ezenwa, 1982) even though problems of acidification and land degradation are associated with it (Ajao, 1993). Similarly at Makali, in northern Sierra Leone, problems have been noted with intense aquaculture in mangrove areas (Johnson and Johnson, 1993). Dams and Barrages Dams and barrages are a major threat to mangrove systems throughout West Africa through their disturbance to the hydrological and tidal regimes of the rivers and through the reduction of sediment supply to deltas. Major dams or barrages currently exist at Diama in Senegal (B and Leung Tuck, 1989), on the Volta at Akossombo in Ghana (Ly, 1980) and at Nangbeto on the Mono River in Togo (CEC, 1987), at Cotonou in Benin, at Buyo in Cte d'Ivoire (Egnankou, 1985), and at Tchimbl and Kingul on the Mbei River as well as at Poubara on the Ogoou River in Gabon. Sayer et al. (1992) report that more than 40 small anti-salt barrages have recently been built in Guinea-Bissau and there are other proposals currently under consideration (e.g. at Onitsja on the Niger River). B and Leung Tuck (1989) have described some of the adverse effects of the anti-salt barrage at Diama on the hydrology of the Senegal River and its estuarine system while others have described the ecological effects of the barrage on the Mono River (Akpagana, 1993). Nevertheless, the reduction in freshwater inflow to the deltas and estuaries of West Africa is likely to become more acute in the near future. Any increased salinity in the deltas is likely to add to the regional stresses due to decreasing rainfall over the last twenty or so years. In Ghana, Ly (1980) found that the reduction of sediment supply via the Volta River initiated the onset of coastal erosion. Oil Pollution Although natural oil seeps have been reported from the mangroves of West Africa (Hutter, 1906), twenty-three out of 62 oilfields with some 1800 oil wells in the Niger Delta occur within the mangroves (CEC, 1987). Oil terminals are situated at Bonny, Brass and Kanuskiri while seismic lines (8,000km x 20-30m wide) and oil pipelines criss-cross the mangroves. Oils spills are common (Snowdon and Ekweozor, 1987) and the seismic lines have allowed saltwater intrusion into brackish and freshwater swamp areas, causing widespread mortality (Fagbami et al., 1988). The relatively minor oil spillage of 250 barrels of Nigerian crude into the Bonny estuary (Snowdon and Ekweozor, 1987) caused the oiling of mangrove prop roots and seedlings and resulted in partial defoliation and death of seedlings within a 500m2 area.

27

An investigation into the causes of vegetation mortality over approximately 200 km2 around the Tsekelewu Oil Field in the Niger delta was carried out (Fagbami et al., 1988) using quantitative photo interpretation of black and white aerial photographs. Soil, surface and ground water samples were taken from the various zones for chemical analyses. Salt concentration decreased progressively landwards, with a marked increase at the salinefresh water transitional zone where the greatest damage occurred. In the transitional saline-fresh water zone, all the vegetation including A. germinans was killed. The massive destruction was attributed to high salinity caused by salt water incursion from the sea as a result of seismic exploration. In Gabon, petroleum exploration has left extensive systems of survey lines through the mangroves which have failed to revegetate since clearing (Lebigre, 1983; J. Fontes, pers. comm.). Similar effects from exploration activities have been reported from the mangroves in Angola (Huntley, 1974). Chronic oil spills have also been noted to have caused the degradation of Mvassa lagoon in the Congo (Camara, 1993) while the oil terminal in the Angolan enclave of Cabinda has been reported to affect the mangroves of Zaire (Sayer et al., 1992). Industrialization and Urbanization The development of large urban centres with significant industrialization poses a number of threats to the mangrove vegetation. Thus, around Abidjan, Cte d'Ivoire, industrial pollution has adversely affected the coastal lagoons (e.g. Lagune Ebri) and their associated mangroves (Carmouze and Caumette, 1985; Zabi, 1982). The decline in water quality has been magnified by sand-extraction from the coastal lagoon system (Egnankou, 1985). Such effects on coastal lagoons are not confined to Cte d'Ivoire but occur, for example, in Ghana (Chemu and Korle lagoons - Biney, 1982), Benin and throughout Nigeria (Ajao and Fagade, 1990). In Gabon (Posso, 1993) reports pollution of the mangroves of the Como River due to the restaurants at Esterias Cape. In Cameroon, the effects from Douala port extension, urban expansion, sand extraction and industrial pollution have caused damage to the mangroves of the Wouri estuary (Din 1991). DDT Fishing Fishing by the use of insecticides is illegal throughout the region but still widely practised, particularly in Cte d'Ivoire (Egnankou, 1985) and in Cameroon (Din 1991). While this activity causes little direct damage to the mangrove vegetation, its adverse effect on any aquatic fauna is likely to cause indirect effects on the vegetation.

28

Weed Invasion Invasion and replacement of native mangrove by the mangrove palm Nypa fruticans also appears to be a local problem. Although known from the fossil record from throughout the Niger delta until 25 m.y. BP (Sowunmi, 1986), the present populations of N. fruticans were introduced to Calabar in 1906 and Oron in 1912 (Wilcox, 1985) from the Singapore Botanic Gardens. This species has spread throughout the Niger, Imo, Bonny and Cross Rivers since then. The spread has been slow but it is perceived to be accelerating over the last few years and it has now reached the Wouri Estuary in Cameroon where its dispersal is facilitated by local villagers who value its thatching properties (Din, 1991; Saenger, pers. obs.). Introduced stands of Nypa have also recently been recorded from the Atlantic coast of Panama, where a similar spread is predicted (Duke, 1991). Coastal Erosion Coastal erosion is widespread in West Africa (Quelennec, 1987; Awosika et al., 1993). However, in Nigeria the erosion of outer barrier beaches in the Niger delta poses the threat of extensive saltwater intrusion into mesohaline mangrove communities (Oyegun, 1990). Management and Rehabilitation Mangrove management and rehabilitation had some early supporters in West Africa. Already in 1909, Rivire described the plan to plant Rhizophora in Tunesia both to provide coastal protection and to produce tan-bark, as ".. vritable hrsie climatologique et culturale". That such mangrove planting was not feasible given the particular climatic conditions, had already been established by Rivire's own culture attempts in Algeria (Grewe 1941) and had already negated the proposal to stabilize the Suez canal banks with mangroves! Similarly by 1911, demands were being made for the protection of mangroves and for the regulation of their exploitation around Douala in Cameroon, in order to safeguard their coastal protective function (Grewe, 1941). In Zaire, tannin extraction dates back to mangrove concessions granted in 1908 and these concessions included (i) riverine buffer zones to prevent erosion and siltation; (ii) concern over traditional user access to the mangroves; and (iii) a condition to replant extracted areas at 1000 seedlings/ha (Pynaert, 1933). The replanted mangroves were reported to have a stem diameter of 15cm after 15 years and to have become mature after 20-25 years (Pynaert, 1933). Despite these early initiatives, mangrove management, conservation and rehabilitation is not generally well established in West Africa - although

29

considerable variation exists in terms of management approaches. The management framework for each of the coastal states is briefly reviewed below: Mauritania: Limited information on the management framework is available although the very limited mangrove relicts of Mauritania are contained within national parks (Table 18). Senegal: A number of national parks containing mangroves have been established (Table 18) and these are managed by the National Park Directorate in the Ministry of Tourism and Nature Conservation. Nonnational park areas are administered by the Direction des Eaux, Forts et Chasses. The Saloum Delta National Park has also been listed as Biosphere Reserve and RAMSAR site since 1984. Nevertheless, mangroves in Senegal remain threatened by the extension of rice cultures and cutting for fuelwood (Diop and B, 1993). Gambia: A number of national parks containing mangroves have been established (Table 18) and are managed by the Wildlife Conservation Department. Attempts to develop sustainable commercial forestry are underway. Guinea-Bissau: There is presently no mangrove management in GuineaBissau although a multidisciplinary coastal ecosystems study is presently underway and legislation for the conservation and management of mangroves is in preparation (Simao, 1993). A number of Ramsar Sites with mangroves exist (Table 18) and the Bijagos Archipelago (11.3N, 16W), with extensive mangrove areas has been proposed as a protected area but its status is uncertain. Other proposed reserves for mangroves include the Cacheu River and Cantanhez Forest (Sayer et al., 1992). Guinea: Concern over mangrove losses has been expressed and a forestry action plan (including mangroves) has been developed which according to Diallo (1993) should provide for sustainable use. Liberia: No mangrove reserves have been established (Sayer et al., 1993) and the mangroves remain under threat from wood exploitation. Sierra Leone: Responsibility for mangrove vegetation is under the Ministry of Agriculture, Forestry and Fisheries (Johnson and Johnson, 1993). Extensive mangroves have been proposed for protection in the Bunce River area (8.50N, 12.83W). Cte d'Ivoire: With the establishment in 1992 of a Direction de l'Environnement within the Ministre de l'Environnement, de la Construction et de l'Urbanisme, studies into the degradation of coastal lagoons are in progress. To date, however, there is no specific mangrove protection (Egnankou, 1993).

30

Ghana: Although possessing an Environment Protection Council since 1974 there are no mangrove reserves and there is no specific mangrove protection. However, management plans have been prepared for some specific wetland areas as RAMSAR sites (Sackey, 1993). Togo: Responsibility for the mangroves falls within the Ministry of Environment and Tourism, specifically the Commission of Forests and Hunting, and the Commission of Flora (Akpagana, 1993). However, there are no existing specific measures for the management of mangroves. Benin: Mangrove management is the responsibility of the Direction de l'Environnement, established in 1991 within the Ministre de l'Environnement, de l'Habitat et de l'Urbanisme. No official protective measures currently exist although a number of informal reserves have been identified. Two types of conservation areas are in use - those of religious significance e.g. the fetish-mangroves on the lots des Avlkt and Mtogbodji, and at Kptou (Blasco 1985; Baglo, 1989) where most exploitative activities are forbidden; and those controlled by local people where exploitative activities are regulated by the village chief. The establishment of biosphere reserves has been in progress since 1981 and a mangrove replanting program has been in place since March 1987 as part of the Beninois-German Co-operative Program "de pche lagunaire", involving 14 villages. In 1987, 12,000 propagules of Rhizophora racemosa were planted near Vekky Dogbodji in Lac Nokou. Since then, planting was extended to Lac Ahm and the coastal lagoon system with 121,000 propagules planted in 1988 and a further 250,000 in 1989. Cost of collection was 0,20FF per plant with the overall cost of planting estimated at $US360/ha (Baglo, 1989). Major problems with these new plantations include (i) prolonged freshwater inundation killing some of the seedlings; (ii) crab predation from Goniopsis pelii; and (iii) insect predation by the larvae of Charaxes castor (Nymphalidae). Nigeria: A Federal Environmental Protection Agency was established in 1988 and, more recently, a National Coastal Management Authority has been proposed to regulate and manage all coastal resources. To date, no mangrove protection is in existence although the Cross River National Park contains some mangrove areas (Table 18). There exists a 247km2 mangrove forest reserve but it is not worked regularly (CEC, 1987). The National Institute of Oceanography and Marine Resources (NIOMR) has recommended that a system of mangrove forest reserves be established to replenish trees that are unavoidably felled by coastal people. It was also suggested that the price of gas also needs to be reduced to an affordable level for coastal communities to reduce the use of mangroves for domestic cooking (Amadi, 1991). Cameroon: The Ministry of Environment was established in 1992 but, at present, there is no legislation to control the management, utilisation or protection of mangroves (Appolinaire, 1993).

31

Equatorial Guinea: Limited information on the management framework is available but, according to Sayer et al. (1992), the mangroves have suffered little degradation although traditionally they have been exploited for firewood and building materials. Gabon: Limited information on the management framework is available and, at present, there are no conservation reserves specifically for mangroves (Sayer et al., 1992). Three Ramsar sites (Table 18) contain some mangrove areas. However, except for around the immediate vicinity of the capital, the mangroves have suffered little degradation due to low population pressure. Congo: Limited information on the management framework is available although at least one mangrove reserve (Conkouati Wildlife Reserve) has been established. Zaire: Limited information on the management framework is available although a 100,000ha Mangrove Reserve (Table 18) was recently established in the Zaire River estuary. Mangrove utilisation leases in Zaire were granted as early as 1908 (Pynaert, 1933) by the Belgian colonial administration. Angola: Limited information on the management framework is available although a large mangrove reserve (Parque Nacional da Qicama) with extensive mangroves has been in existence since 1957. With the unrest in the country, this reserve has been threatened by illegal grazing, cotton production, oil exploration and extraction as well as diamond prospecting (Huntley, 1974). Major Data Gaps The aim of this review was to assemble and evaluate existing information about the mangrove vegetation of the Atlantic coast of Africa and, in doing so, it has identified a number of areas where existing knowledge is extremely limited or lacking completely. By identifying those areas, this review might stimulate or encourage studies to fill those needs and thereby enhance the understanding of mangrove ecology on the African Atlantic coast and facilitate the conservation and management of these mangrove systems on a sustainable basis. One of the obvious information gaps is the absence of detailed distributional and abundance data of mangrove species and their associates. Existing data has been summarised in Table 6 but inadequate data exist for several areas including Equatorial Guinea, Congo, Togo and Liberia. With reliable distribution and abundance data, additional questions may be answered. For instance, why is R. mangle apprently absent from the drier coasts of Ghana, Cte d'Ivoire and Benin while present in Senegal? It is noteworthy that R. mangle occurs abundantly in Florida and Yucatan, Mexico with a similar rainfall to these West African areas. Are the West

32

African mangrove populations of this species less adapted to deal with high salinities as suggested by the data in figure 9? In turn, such investigations may throw light on whether or not there has been some genetic differentiation between the mangrove species of the Atlantic coasts of Africa and America. Why do they appear not to have segregated, at least morphologically, during a separation of 10 million or so years? Has there been biochemical or physiological differentiation? More intriguingly, is there genetic leakage across the Atlantic? At present, there is no systematic study of the phenology of the mangroves of the Atlantic coast of Africa; the summarized information has been extracted from herbarium data concerning individual specimens and from incidental observations. The absence of systematic data makes comparisons with the Americas difficult. More importantly, it is difficult to predict when flowers and fruits are available; this is particularly significant for ecological studies of those species (e.g. Rhizophora) where reliable species identification requires the presence of flowers and fruits. Similarly, there are no litter fall data for the Atlantic coast of Africa and only very limited biomass data. While data for many of the individual species are available from the American Atlantic coast, the absence of these data from the African coast makes functional comparisons between the African mangroves and those of other areas difficult. Litter and wood productivity estimates are important respectively for quantifying the overall contributions made by mangroves to nearshore waters and for assessing the potential for commercial wood production from mangrove systems. There appears to be little effort put into the establishment of sustainable forestry operations of the mangroves as have been developed in India, Bangladesh, Malaysia and Thailand. There is little doubt that at least the mangroves of Nigeria, Cameroon, Gabon and Equatorial Guinea could be exploited on a sustainable basis. Similarly, with the exception of Benin, there have been few attempts at mangrove afforestation not only to produce wood but to enhance the coastal stability of the coastal regions - as, for example, in Bangladesh (Saenger and Siddiqi, 1993). In the case of the Niger delta, efforts at developing a sustainable mangrove forestry industry would appear to be a prioroity. From a management perspective, there have been relatively few hydrological studies on the effects of dams and barrages on the downstream mangrove vegetation. Such studies are vital not only for the continued functioning of the mangrove systems but also in terms of the supply of sediments to coastal areas - a factor already implicated in coastal erosion. Similarly, the likely effects on coastal erosion, particularly in relation to mangroves, need to be investigated. Although the extent of acid-sulphate soils is reasonably well-known (Table 17), their pedogenesis is only now being understood. The role of mangrove

33

vegetation in the genesis and development of acid-sulphate soils is still in need of further investigation (Feller et al., 1989), particularly in view of the widespread and on-going conversion of mangrove soils to rice cultivation. Finally, given the mangrove areas, the population pressures and the per capita gross national product (GNP) of the west African nations (Table 8), it would seem that any conservation initiatives would be most rewarding in Gabon, Congo, Cameroon and Senegal in that order. On the other hand, conservation initiatives appear to be most urgently needed in Guinea, Sierra Leone, Cte d'Ivoire, Benin, Nigeria, Cameroon and Equatorial Guinea. Acknowledgements This review was prepared while the author was on study leave at the Institut de la Carte Internationale de la Vgtation (C.N.R.S. - U.P.S.), Toulouse, France and it is a pleasure to thank the Director, Prof. F. Blasco, for making the Institute's facilities available and for many inspiring discussions during my visit. In addition, I gratefully acknowledge the assistance of the Institute's librarian, Mrs. M. Vassal for her untiring efforts in chasing obscure publications on my behalf, and Miss H.F. Bellan, who taught me some of the finer points of image analysis. I thank UNIDO for providing the opportunity to study the mangroves of West Africa and Prof. F. Blasco, Prof. D. McConchie and Dr. A. Specht for their constructive comments on earlier drafts of this review.

34

References Cited
Adam, J-G., 1958. Floristique des paturages sals (halophytes et subhalophytes) et vgtation des rizires du sine-Saloum (Sngal). Journ. d'Agric. Trop. et de Botan. Appl. 5:505-541. Adam, J-G., 1965. La vgtation du delta du Sngal en Mauritanie - le cordon littoral et l'le de Thiong. Bull. de l'I.F.A.N. 27:121-138. Adam, J-G., 1970a. La vgtation du cordon littoral et lagunaire du Cap des Palmes (Cap Palmas) (Libria). Bull. Soc. bot. Fr. 117:419-428. Adam, J-G., 1970b. La vgtation littorale aux environs de Buchanan (Libria). Bull. de l'I.F.A.N. 32:995-1018. Adams, C.D., 1957. Observations on the fern flora of Fernando Po. I. A description of the vegetation with particular reference to the Pteridophyta. J. Ecol. 45:479-494. Adegbehin, J.O., 1993. A synthesis of the mangroves of Nigeria. In: Diop, E.S., C.D. Field and M. Vannucci (Eds.), Proceedings of a Workshop on Conservation and Sustainable Utilization of Mangrove Forests in Latin America and Africa Regions. Dakar, 20-22 January, 1993. ITTO/ISME Project PD114/90 (F), pp. 12-13. Airy-Shaw, H.K., 1947. The vegetation of Angola. J. Ecol. 35:23-48. Ajao, E.A., 1993. Mangrove ecosystems in Nigeria. In: Diop, E.S., C.D. Field and M. Vannucci (Eds.), Proceedings of a Workshop on Conservation and Sustainable Utilization of Mangrove Forests in Latin America and Africa Regions. Dakar, 20-22 January, 1993. ITTO/ISME Project PD114/90 (F), pp. 13-14. Ajao, E.A. and S.O. Fagade, 1990. A study of the sediments and communities in Lagos Lagoon, Nigeria. Oil & Chem. Pollution 7:85-117. Akpagana, K., 1993. The mangrove ecosystems and associated formations in the south-east of Togo. In: Diop, E.S., C.D. Field and M. Vannucci (Eds.), Proceedings of a Workshop on Conservation and Sustainable Utilization of Mangrove Forests in Latin America and Africa Regions. Dakar, 20-22 January, 1993. ITTO/ISME Project PD114/90 (F), pp. 1011. Amadi, A.A., 1990. A comparative ecology of estuaries in Nigeria. Hydrobiologia 208:27-38. Amadi, A.A., 1991. The coastal and marine environment of Nigeria - aspects of ecology and management. Nigerian Institute for Oceanography and Marine research, Tech. Paper No. 76, pp. 34. Amlalo, D.S., 1990. Tourism and ecological balance in an urban situation: a case study of Labadi Pleasure Beach area. Unpubl. thesis, University of Salford, Manchester. Amobi, C.C., 1974. Periodicity of wood formation in twigs of some tropical trees in Nigeria. Ann. Bot. 38:931-936. Anthony, E., 1989. Chenier plain development in northern Sierra Leone, West Africa. Mar. Geol. 90:297-309. Appolinaire, Z., 1993. The mangrove swamps of Cameroon. In: Diop, E.S., C.D. Field and M. Vannucci (Eds.), Proceedings of a Workshop on Conservation and Sustainable Utilization of Mangrove Forests in Latin America and Africa Regions. Dakar, 20-22 January, 1993. ITTO/ISME Project PD114/90 (F), pp. 16-18. Assemien, P., 1969. Pollen fossile de Rhizophora Bogu (Basse valle du Sngal). Pollen et Spores 11:73-81. Assemien, P., 1971. Etude comparative des flores actuelles et quaternaires recent de quelques paysages vegetaux d'Afrique de l'Ouest. Unpubl. Thesis, University of Abidjan, Abidjan. Attims, Y. and G. Cremers, 1967. Les radicelles capillaires des paltuviers dans un mangrove de Cte d'Ivoire. Adansonia (Ser. 2) 7:547-551. Awosika, L.F., A.C. Ibe and P. Shroader (Eds.), 1993. Coastlines of Western Africa. American Society of Civil Engineers, New York. B, M. and K.D. Leung Tuck, 1989. First effects of the Diama's dam on the hydrological evolution in Senegal estuary, West Africa. Proceedings Sixth Symposium on Coastal and Ocean Management/ASCE, Charleston, South Carolina, pp. 1875-1884. Baglo, M., 1989. La mangrove du Bnin: grands equilibres cologique et perspectives d'amnagement. Unpubl. Thesis, Paul Sabatier University, Toulouse. Baillaud, E., 1904. La question des paltuviers. Journ. d'Agr. trop. 37:200-206. Barth, H., 1982. The biogeography of mangroves. In: Sen, D.N. and K.S. Rajpurohit (Eds.), Contributions to the Ecology of Halophytes, Tasks for Vegetation Science, Vol. 2, pp. 85110, W. Junk, The Hague.

35 Barusseau, J-P., E.H.S. Diop, P. Giresse, J. Monteillet and J-L. Saos, 1986. Consequences sedimentologiques de l'evolution climatique fini-Holocne dans le delta du Saloum (Sngal). Oceanogr. Trop. 21:89-98. Bertrand, F., 1991. Contribution l'tude de l'environnement et de la dynamique des mangroves de Guine: donnes de terrain et apport de la tldtection. Unpubl. Thesis, Universit de Bordeaux. Bessedik, M., 1981. Une mangrove Avicennia L. en Mditerrane occidentale au Miocne infrieur et moyen. Implications palogographiques. C.R. Acad. Sc. (Paris) 293:469-472. Bessedik, M., 1985. Rconstruction des environnments miocnes des regions nord-ouest mditerranenes partir de la palynologie. Unpubl. Thesis, L'Universit des Sciences et Techniques du Languedoc, Montpellier. Biney, C.A., 1982. Preliminary survey of the state of pollution of the coastal environment of Ghana. Oceanol. Acta, Actes Symposium International sur les lagunes ctires, SCOR/IABO/UNESCO, Bordeaux, 8-14 September 1981, pp. 39-43. Blasco, F., 1983. Mangroves du Sngal et de Gambie. CNRS/Paul Sabatier University, Toulouse, 86 pp. Blasco, F., 1985. Mangroves du Bnin. CNRS/Paul Sabatier University, Toulouse, 56 pp. Boughey, A.S., 1957. Ecological studies of tropical coastlines. I. The Gold Coast, West Africa. J. Ecol. 45:665-687. Boy, M., F. Baltzer, C. Caratini and J.F. Villiers, 1975. Mangrove of the Wouri Estuary, Cameroon. In: Proc. Intern. Symp. Biol. & Management of Mangroves, Vol.2: 431-455. Breteler, F.J., 1969. The Atlantic species of Rhizophora. Acta Bot. Neerl. 18:434-441. Camara, E., 1993. Mangroves of Congo. In: Diop, E.S., C.D. Field and M. Vannucci (Eds.), Proceedings of a Workshop on Conservation and Sustainable Utilization of Mangrove Forests in Latin America and Africa Regions. Dakar, 20-22 January, 1993. ITTO/ISME Project PD114/90 (F), pp. 20-21. Carmouze, J-P. and P. Caumette, 1985. Les effets de la pollution organique sur les biomasses et activits du phytoplankton et des bacteries htrotrophes dans la lagune Ebri (Cte d'Ivoire). Revue Hydrobiologique Tropicale 18:183-211. Chapman, V.J., 1977. Introduction. In: Chapman, V.J. (Ed.), Ecosystems of the World. 1. Wet coastal ecosystems. Elsevier Sci. Publ. Co., Amsterdam, pp; 1-29. Chapman, V.J., 1944. The 1939 Cambridge University Expedition to Jamaica. I. A study of the botanical processes concerned in the development of the Jamaican shoreline. Bot. J. Linn. Soc. 52:407-447. Commission of the European Communities, 1987. Mangroves d'Afrique et de Madagascar. Protection et mise en valeur (Vol. 1-3). Leiden Univ. (Netherlands), Centre for Environmental Studies in cooperation with Societ d'Ecoamnagement. Compre, P., 1963. The correct name of the Afro-American black mangrove.. Taxon 12:150152. Darteville, E., 1950. Les mangroves du Congo et les autres mangrove d'Afrique occidentale. Bull. Sance Inst. Roy. Colon. Belge 21:946-972. Delevoy, G., 1945. Les mangroves africaines. Bull. Soc. Central Forest. Belgique 52:84-89; 129-130. Delibrias, G., L. Ortlieb and N. Petit-Maire, 1976. New C14 dates for the Atlantic Sahara. Tentative interpretations. J. Human Evol. 5:535-546. Dent, J.M., 1947. Some soil problems in empoldered rice lands in Sierra Leone. Empire J. Exp. Agr. 15:206-212. Diallo, A., 1993. The mangrove swamps of Guinea. In: Diop, E.S., C.D. Field and M. Vannucci (Eds.), Proceedings of a Workshop on Conservation and Sustainable Utilization of Mangrove Forests in Latin America and Africa Regions. Dakar, 20-22 January, 1993. ITTO/ISME Project PD114/90 (F), pp. 6-7. Din, N., 1991. Contribution l'tude botanique et cologique des mangroves de l'estuaire du Cameroun. Unpubl. Thesis, Universit de Yaound, Yaound. Diop, S., 1990. La cte ouest-Africaine du Saloum (Sngal) la Mellacore (Rp. de Guine). Institut Francais de Recherche Scientifique pour le dveloppement en coopration. Collection Etudes et Thses, ORSTOM, Paris, pp. 379. Diop, E.S. and M. B, 1993. The mangrove swamps of Senegal and Gambia. In: Diop, E.S., C.D. Field and M. Vannucci (Eds.), Proceedings of a Workshop on Conservation and Sustainable Utilization of Mangrove Forests in Latin America and Africa Regions. Dakar, 20-22 January, 1993. ITTO/ISME Project PD114/90 (F), pp. 3-4.

36 Diop, E.S., C.D. Field and M. Vannucci (Eds.), 1993. Proceedings of a Workshop on Conservation and Sustainable Utilization of Mangrove Forests in Latin America and Africa Regions. Dakar, 20-22 January, 1993. ITTO/ISME Project PD114/90 (F). Dost, H., (Ed.), 1988. Selected papers of the Dakar Symposium on acid sulphate soils. Dakar, January 1986. ILRI, P.O. Box 45, 6700 AA Wageningen, pp. 251. Drabble, E., 1908. The bark of the red and white mangroves. Q. J. Inst. Comm. Res. Trop. 3:33-39. Drabble, E. and M. Nierenstein, 1907. A note on the West-African mangroves. Q. J. Inst. Comm. Res. Trop. 2:15. Duke, N.C., 1991. Nypa in the mangroves of Central America: introduced or relict? Principes 35:127-132. Durand, J.R. and J.M. Chantraine, 1982. L'environnement climatique des lagunes ivoriennes. Revue Hydrobiologie Tropicale 15:85-113. Durand, J.R. and M. Skubich, 1982. Les lagunes ivoriennes. Aquaculture 27:211-250. Edwards, A.W.A. and O.E.C. Ekundayo, 1982. Biomass analysis of mangrove swamp successional species. In: Gopel, B., R.E. Turner, R.G. Wetzel and D.F. Whigham (Eds.), Wetlands: Ecology and Management. National Institute of Ecology and International Scientific Publications, New Delhi, 7:1-10. Egnankou, W.M., 1985. Etude des mangroves de Cte d'Ivoire: Aspect cologique et rcherches sur les possibilities de leur amnagement. Unpubl. Thesis, Paul Sabatier University, Toulouse. Egnankou, W.M., 1993. The mangrove swamps of Cte d'Ivoire. In: Diop, E.S., C.D. Field and M. Vannucci (Eds.), Proceedings of a Workshop on Conservation and Sustainable Utilization of Mangrove Forests in Latin America and Africa Regions. Dakar, 20-22 January, 1993. ITTO/ISME Project PD114/90 (F), p. 9. Emery, K.O. and D.G. Aubrey, 1991. Sea levels, land levels and tide gauges. Springer Verlag, New York. Elouard, P., 1968. Le Nouakchottien, tage quaternaire de Mauretanie. Ann. Fac. Sci., Dakar 22:121-137. Exell, A.W., 1945. Catalogue of the vascular plants of S. Tom (with Principe and Annobon). British Museum of Natural History, London. Exell, A.W., 1973. Angiosperms of the islands of the Gulf of Guinea (Fernando Po, Sao Tom, Annobon). Bull. Br. Mus. Nat. Hist., Botany 4:325-411. Fagbami, A.A., E.J. Udo and C.T.I. Odu, 1988. Vegetation damage in an oil field in the Niger Delta of Nigeria. J. Trop. Ecol. 4(1):61-75. Faure, H. and L. Hebrard, 1977. Variations des lignes de rivage au Sngal et en Mauretanie au cours de l'Holocne. Studia Geologica Polonica 52:143-157. Feller, C., J. Trichet, J.C. Fontes and C. Marius, 1989. Sur le rle de la vgtation dans le stockage du soufre dans les sols de mangrove (Sngal). Resultats preliminaires. Soil Biol. Biochem. 21(7):947-952. Gaillard, M., M. Gaillard, J. Lang and J. Lucas, 1982. Etude sdimentologique des milieux de mangrove du Bnin occidental (Afrique de l'Ouest). Oceanol. Acta, Actes Symposium International sur les lagunes ctires, SCOR/IABO/UNESCO, Bordeaux, 8-14 September 1981, pp. 129-137. Giglioli, M.E.C. and D.F. King, 1966. The mangrove swamps of Keneba, Lower Gambia River basin. III. Seasonal variations on the chloride content of swamp soils, with observations on the water level and chloride concentration of free soil water under a barren mud flat during the dry season. J. Appl. Ecol. 3:1-19. Giglioli, M.E.C. and I. Thornton, 1965. The mangrove swamps of Keneba, Lower Gambia River basin. I. Descriptive notes on the climate, the mangrove swamps and the physical composition of their soils. J. Appl. Ecol. 2:81-103. Gledhill, D., 1963. The ecology of the Aberdeen Creek mangrove swamp. J. Ecol. 51:694703. Gowthorpe, P. and B. Lamarche, 1993. Les mangroves de la Mauritanie. In: Diop, E.S., C.D. Field and M. Vannucci (Eds.), 1993. Conservation et utilisation rationelle des frets de mangrove de l'Amerique Latine et de l'Afrique. Dakar, 20-22 January, 1993. ITTO/ISME Project PD114/90 (F), pp. 3-21. Gossweiler, J. and F.A. Mendona, 1939. Carta Fitogeogrfica de Angola. Repblica Portuguesa Ministrio das Colnias, Lisbon. Grewe, F., 1941. Afrikanische Mangrovelandschaften. Verbreitung und wirtschaftsgeographische Bedeutung. Wiss. Verff. Deutsch. Museum f. Lnderkunde (Leipzig) N.F. 9:105-177.

37 Hachimou, I., 1993. The mangrove swamps of Benin. In: Diop, E.S., C.D. Field and M. Vannucci (Eds.), Proceedings of a Workshop on Conservation and Sustainable Utilization of Mangrove Forests in Latin America and Africa Regions. Dakar, 20-22 January, 1993. ITTO/ISME Project PD114/90 (F), pp. 11-12. Hart, M.G.R., A.J. Carpenter and J.W.O. Jeffrey, 1963. Problems in reclaiming saline mangrove soils in Sierra Leone. Agron. Trop. 18:800-802. Healey, M.J., R.A. Moll and C.O. Diallo, 1988. Abundance and distribution of bacterioplankton in the Gambia River, West Africa. Microb. Ecol. 16(3):291-310. Hdin, L., 1928. L'exploitation du paltuvier dans la baie de Manoka (Cameroun). Rvue de Botanique applique 8:623-626. Hdin, L., 1933. Observations sur la vgtation des bords lagunaires dans la rgion de Grand-Bassam et de Bingerville (Cte d'Ivoire). La Terre et la Vie 3:596-609. Heske, F., 1937. Kolonialwald und Kolonialwaldwirtschaft. D. Forstb.-Zeit. 3:632-657. Hesse, P.R., 1961. Differences between the soils of the Rhizophora and Avicennia mangrove swamps in Sierra Leone. Plant and Soil 14:335-346. Huntley, B.J., 1974. Outlines of wildlife conservation in Angola. Jl. Southern Africa Wildlife Management Association 4:157-166. Hutchings, P. and P. Saenger, 1987. Ecology of Mangroves, Queensland University Press, Brisbane. Hutchinson, J., J.M. Dalziel, R.W.J. Keay and F.N. Hepper, 1954/72. Flora of Tropical West Africa. 2nd Edition, 3 volumes. Hutter, H., 1906. Im lgebiet von Kamerun. Globus 89:1-7. IUCN, 1994. 1993 United Nations list of national parks and protected areas. Prepared by WCMC and CNPPA. IUCN, Gland, Switzerland and Cambridge, UK, pp. 315. Isebor, C.E., 1993. Dynamics of mangrove ecosystem along the Nigerian coastline. In: Awosika, L.F., A.C. Ibe and P. Shroader (Eds.), Coastlines of Western Africa. American Society of Civil Engineers, New York, pp. 209-219. Isebor C.E. and L.F. Awosika, 1993. The Nigerian mangrove resources; status and management. In: Diop, E.S., C.D. Field and M. Vannucci (Eds.), Proceedings of a Workshop on Conservation and Sustainable Utilization of Mangrove Forests in Latin America and Africa Regions. Dakar, 20-22 January, 1993. ITTO/ISME Project PD114/90 (F), pp. 14-16. Jackson, G., 1964. Note on West African vegetation. I. Mangrove vegetation at Ikorodu, Western Nigeria. Jl. West African Science Association 9(2):98-111. Jackson, G., 1985. A West African mangrove swamp bibliography. In: Wilcox, B.H.R. and C.P. Powell (Eds.), The mangrove ecosystem of the Niger delta: Proceedings of a workshop., University of Port Harcourt, Port Harcourt, pp. 280-307. Jacques-Felix, H., 1957. Les Rhizophora de la mangrove atlantique d'Afrique. J. Agric. Trop. Bot. Appl. 4:343-347. Jacques-Felix, H. and R. Chezeau, 1960. Sols et groupements vgtaux de la zone littorale de Guine dans leurs rapports avec la riziculture. Agron. Trop. 15:325-341. Jenk, J., 1970. Root system of tropical trees 5. The peg-roots and the pneumathodes of Laguncularia racemosa Gaertn. Preslia (Prague) 42:105-113. Jimenez, J.A., 1987. A clarification on the existence of Rhizophora species along the Pacific coast of Central America. Brenesia 28:25-32. John, D.M. and G.W.A. Lawson, 1990. Review of mangrove and coastal ecosystems in West Africa and their possible relationships. Est. Coast. Shelf Sci. 31(5):505-518. Johnson, R. and R. Johnson, 1993. Mangroves of Sierra Leone. In: Diop, E.S., C.D. Field and M. Vannucci (Eds.), Proceedings of a Workshop on Conservation and Sustainable Utilization of Mangrove Forests in Latin America and Africa Regions. Dakar, 20-22 January, 1993. ITTO/ISME Project PD114/90 (F), pp. 7-9. Jordan, H.D., 1963. The vegetation of mangrove swamps in West Africa. Agro. Trop. 18:796797. Jordan, H.D., 1964. The relation of vegetation and soil to development of mangrove swamps for rice growing in Sierra Leone. Jl. Appl. Ecol. 1:209-212. Kalck, Y., 1978. Evolution des zones mangroves du Sngal au Quaternaire rcent: Etudes gologiques et gochimiques. Unpubl. PhD Thesis, University of Louis Pasteur, Strasbourg. Keay, R.W.J., 1953. Rhizophora in West Africa. Kew Bull. 8:121-127. Kinako, P.D.S., 1977. Conserving the mangrove forest of the Niger Delta. Biol Conserv. 11:35-39. Kingsley, M., 1897. Travels in West Africa. Macmillan Co. Ltd., London.

38 Kunkel, G., 1966. ber die Struktur und Sukzession der Mangrove Liberias und deren Randformationen. Ber. Schweiz. Bot. Ges. 75:20-40. Kwei, E., 1977. Biological, chemical and hydrological characters of coastal lagoons of Ghana, West Africa. Hydrobiologia 56:157-174. Lang, J. and G. Paradis, 1977. Un example d'environnement sdimentaire, biodtritique, non carbonat marin et continental, holocne en climat intertropical: le domaine margino-littoral du Bnin (ex-Dahomey). Rev. Gogr. Phys. Gol. Dynam. 19:295-312. Lang, J. and G. Paradis, 1984. Le Quaternaire margino-littoral bninois (Afrique de l'Ouest). Synthse des datations au carbone 14. In: Coetzee, J.A. and E.M. van Zinderen Bakker (Eds.), Palaeoecology of Africa and the surrounding islands. Balkema, Rotterdam, pp; 6576. Lawson, G.W., 1966. The littoral ecology of West Africa. Oceanogr. Mar. Biol. Ann. Rev. 4: 405-448. Lebigre, J-M., 1983. Les mangroves des rias du littoral gabonais. Essai de cartographie typologique. Rev. Bois et Forts des Tropiques 199:3-28. Lebrun, J., 1954. Sur la vgtation du secteur littoral du Congo belge. Vegetatio 5-6:157160. Lezines, A.M., 1988. New pollen data from the Sahel, Senegal. In: Quaternary Palynology of Tropical Areas. Rev. Palaebot. Palynol. 55:141-154 (special issue edited by Ward JV and Liu K-B) Longhurst, A.R., 1962. A review of the oceanography of the Gulf of Guinea. Bull. de l'I.F.A.N. 24:633-663. Ly, C.K., 1980. The role of the Akosombo Dam on the Volta River in causing coastal erosion in central and eastern Ghana (West Africa). Mar. Geol. 37:323-332. Marius, C., 1979. Les mangroves du Sngal. ORSTOM, Dakar. Marius, C., 1981. Ecologie des mangroves du Sngal. In: Les cosystemes ctiers de l'Afrique de l'Ouest: lagunes, stuaires et mangroves. Rapport d'un Atelier, Dakar, 11-15 Juin 1979.; Rapp. UNESCO Sci. Mer. 17. pp. 43-44. Marius, C., 1984. Contribution l'tude des mangroves du Sngal et de la Gambie. Ecologie - pedologie - gochimie - mise en valeur et amenagement. Unpubl. Thesis, University of Louis Pasteur, Strasbourg. Marius, C. and J. Lucas, 1982. Evolution gochemique et exemple d'amnagement des mangroves au Sngal (Casamance). Oceanol. Acta, Proc. Intern. Symp. on coastal lagoons, SCOR/IABO/UNESCO, Bordeaux, France, 8-14 September, 1981, pp. 151-160. Michel, P. and P. Assemien, 1970. Etudes sedimentologiques et palynologiques des sondages de Bogu (Basse valle du Sngal) et leur interpretation morphoclimatique. Rv. Gomorph. dyn. 19:98-113. Naurois, R. de and F. Roux, 1965. Les mangroves d'Avicennia les plus septentrionales de la cte occidentales d'Afrique. Bull. de l'I.F.A.N. 27:842-854. Nicou, R., 1956. Prsence du Laguncularia racemosa dans la mangrove du pseudo-delta du Sngal. Notes Africaines 71:67-68. Ntiamoa-Baidu, Y., 1991. Seasonal changes in the importance of coastal wetlands in Ghana for wading birds. Biol. Cons. 57:139-158. Oyegun, C.U., 1990. The management of coastal zone erosion in Nigeria. Ocean & Shoreline Managt. 14:215-228. Paradis, G., 1976. Contribution l'tude de la flore et de la vgtation littorales du Dahomey. Bull. du Mus. National Hist. Naturelle, Paris (serie 3), 383:33-67. Paradis, G., 1979. Observations sur Laguncularia racemosa Gaertn. et Dalbergia ecastaphyllum (L.) Taub. dans la successions secondaires de la mangrove du Bnin. Bull. de l'I.F.A.N. 41:92-102. Paradis, G., 1980. Un cas particulier de zones dnudes dans les mangroves d'Afrique de l'Ouest: celles dues l'extraction de sel. Bull. du Mus. National Hist. Naturelle, Paris (serie 4), 2:227-261. Paradis, G., 1981. Ecologie et gomorphologie littorale en climat subquatorial sec: la vgtation ctire du bas Bnin occidental. Ann. Univ. Abidjan (ser. E) 14:7-56. Paradis, G., 1989. Observations sur la dgradation d'origine anthropique des mangroves de Fresco et de Grand Bassam (Cte d'Ivoire). Candollea 44:453-483. Paradis, G. and E. Adjanohoun, 1974. L'impact de la fabrication du sel sur la vgtation de mangrove et la gomorphologie dans le Bas-Dahomey. Ann. Univ. Abidjan, Series E, 7:599-612. Pellegrin, F., 1952. Les Rhizophoraces de l'Afrique quatoriale franaise. Notulae Systematicae 14:292-300.

39 Posso, P., 1993. The mangrove swamps of Gabon. In: Diop, E.S., C.D. Field and M. Vannucci (Eds.), Proceedings of a Workshop on Conservation and Sustainable Utilization of Mangrove Forests in Latin America and Africa Regions. Dakar, 20-22 January, 1993. ITTO/ISME Project PD114/90 (F), pp. 18-20. Pynaert, L., 1933. La mangrove congolaise. Bull. Agro. Congo Belge. 23:184-207. Quelennec, R.E., 1987. L'rosion sur les ctes Ouest Africaines. Nature et Ressources 23:4. Rabinowitz, D., 1978. Dispersal properties of mangrove propagules. Biotropica 10:47-57. Rivire, C., 1909. Les vgtaux tannifres dans le nord d'Afrique. Journ. d'Agr. trop. 94:101103. Robinson, S.P. and G.P. Jones, 1986. Accumulation of glycinebetaine in chloroplasts provides osmotic adjustment during salt stress. Aust. J. Pl. Physiol. 13:659-668. Robyns, W., 1950. Botanique du Congo Belge II. La flore. III. La vgtation. IV. Les territoires phytogographiques. Extract de l'Encyclopdie du Congo Belge, Editions Bieleveld, Brussels, pp. 390-424. Rosevear, D.R., 1947. Mangrove swamps. Farm and Forest 8:23-30. Rossignol-Strick, M. and D. Duzer, 1979. West African vegetation and climate since 22,500 BP from deep-sea core palynology. Pollen et Spores 21:105-134. Sackey, I., 1993. State of mangroves in Ghana. In: Diop, E.S., C.D. Field and M. Vannucci (Eds.), Proceedings of a Workshop on Conservation and Sustainable Utilization of Mangrove Forests in Latin America and Africa Regions. Dakar, 20-22 January, 1993. ITTO/ISME Project PD114/90 (F), p. 10. Saenger, P. and J. Moverley, 1985. Vegetative phenology of mangroves along the Queensland coastline. Proc. Ecol. Soc. Aust. 13:257-265. Saenger, P. and S.C. Snedaker, 1993. Pantropical trends in mangrove above-ground biomass and annual litter fall. Oecologia 96:293-299. Saenger, P. and N.A. Siddiqi, 1993. Land from the sea: the mangrove afforestation program of Bangladesh. Ocean and Coastal Management 20:23-39. Saenger, P., E.J. Hegerl and J.D.S. Davie (Eds.), 1983. Global Status of Mangrove Ecosystems. The Environmentalist 3 (Supplement):1-88. Sayer, J.A., C.S. Harcourt and N.M. Collins (Eds.), 1992. The conservation atlas of tropical forests - Africa. Macmillan Publishing Ltd., London. Savory, H.J., 1953. A note on the ecology of Rhizophora in Nigeria. Kew Bull. 8:127-128. Schnell, R., 1952. Contribution une tude phytosociologique et gographique de l'Afrique occidentale: les groupements et les units gobotaniques de la rgion guinenne. Mm. de l'I.F.A.N. 18:14-234. Schwarz, B., 1991. Identification, establishment and management of specially protected areas in the WACAF Region. Draft report by IUCN for UNEP. Simao, A., 1993. The mangrove swamps of Guinea-Bissau. In: Diop, E.S., C.D. Field and M. Vannucci (Eds.), Proceedings of a Workshop on Conservation and Sustainable Utilization of Mangrove Forests in Latin America and Africa Regions. Dakar, 20-22 January, 1993. ITTO/ISME Project PD114/90 (F), pp. 5-6. Snedaker, S.C., 1982. Mangrove species zonation: Why? In: Contributions to the Ecology of Halophytes (Eds. D.N. Sen and K.S. Rajpurohit), Tasks for Vegetation Science, Vol. 2, pp. 111-125, W. Junk, The Hague. Snowden, R.J. and I.K.E. Ekweozor, 1987. The impact of a minor oil spillage in the estuarine Niger Delta. Mar. Pollut. Bull. 18:595-599. Sowunmi, M.A., 1981. Late Quaternary environmental changes in Nigeria; Pollen et Spores 23:125-148. Sowunmi, M.A., 1986. Change of vegetation with time. In: Lawson, G.W. (ed.), Plant ecology in West Africa. John Wiley & Sons, Chichester, pp. 273-307. Tastet, J.P., 1981. Morphologie des littoraux sdimentaires lies aux variations du niveau de la mer: exemple du golfe de Guine. Oceanis 7:455-472. Thompson, R.C.M., 1945. Studies on the breeding places and control of Anopheles gambiae and A. gambiae var. melas in coastal districts of Sierra Leone. Bull. Entomol. Res. 36:185-252. Thornton, I. and M.E.C. Giglioli, 1965. The mangrove swamps of Keneba, Lower Gambia River basin. II. Sulfur and pH in the profiles of swamp soils. J. Appl. Ecol. 2:257-269. Tomlinson, P.B., 1986. The Botany of Mangroves. Cambridge University Press. Tomlinson, T.E., 1957. Relationship between mangrove vegetation, soil texture and reaction of surface soil after empoldering saline swamps in Sierra Leone. Trop. Agric. 34:41-50. Trochain, T.L. and L. Dulau, 1942. Quelques particularits d'Avicennia nitida (Verbenaces) de la mangrove ouest-africaine. Bull. Soc. Hist. Nat. Toulouse 77:271-281.

40 Ukpong, I.E., 1989. An evaluation of the ecological status and environmental relations of mangrove swamps in south-eastern Nigeria. Unpubl. Thesis, University of Ibadan, Ibadan. Ukpong, I.E., 1992. The structure and soil relations of Avicennia mangrove swamps in southeastern Nigeria. Trop. Ecol. 33:1-16. UNESCO, 1979. Map of the world distribution of arid regions: explanatory note. MAB Technical Notes 7, pp. 54. UNESCO, 1983. Atelier d'tude des mangroves au Sud de l'estuaire du Saloum: DiombossBandiale (Sngal). (Final Report) UNESCO Division des Sciences de la Mer, Dakar. Vanden Berghen, C., 1984. L'extraction du sel en Basse-Casamance (Sngal) et les ffets de cette industrie traditionnelle sur la formation des paysages. Journ. d'Agric. Trad. et de Botan. Appl. 31:235-244. Verstraete, J.M., 1989. Le niveau de la mer le long des ctes de l'Afrique de l'Ouest et l'quateur. Hausse probable du niveau marin l'chelle sculaire. COMARAF Publ. Ser. documentaire No. 4, Dakar, 5-43. Villiers, J-F., 1973. Etude floristique et phytosociologique d'une mangrove atlantique sur substrat rocheux du littoral gabonais. Ann. Fac. Sci. Cameroun 14:3-46. Walter, H. and H. Lieth, 1960. Klimadiagram Weltatlas. G. Fischer Verlag, Jena. Walter, H and M. Steiner, 1936/7. Die kologie der Ost-Afrikanischen Mangroven. Z. Bot. 30:65-193. Wilcox, B.H.R., 1985. Angiosperm flora of the mangrove ecosystem of the Niger Delta. In: Wilcox B.H.R. and C.P. Powell (Eds.), The mangrove ecosystem of the Niger delta: Proceedings of a workshop. University of Port Harcourt, Port Harcourt, pp. 34-44. Wokoma, S.A. and B.I. Ezenwa, 1982. Construction of fish ponds in the mangrove swamps of the Niger Delta. Technical Paper No. 9, Nigerian Institute for Oceanography and Marine Research. Victoria Island Lagos. Wyn Jones, R.B. and R. Storey, 1981. Betaines. In: Paleg, L.G. and D. Aspinall (Eds.), Physiology and biochemistry of drought resistance in plants. Academic Press, Sydney, pp. 171-204. Zabi, S.G., 1982. Les peuplements benthiques lagunaires lis la pollution en zone urbaine d'Abidjan (Cte d'Ivoire). Oceanol. Acta, Proc. Intern. Symp. on coastal lagoons, SCOR/IABO/UNESCO, Bordeaux, France, 8-14 September, 1981, pp. 441-455.

41

Table 1. Spring tidal ranges at a number of western African localities.

Location Cape Bimbia, Cameroon Escravos, Nigeria Pointe Olga, Cameroon Takoradi, Ghana Dakar, Senegal Cotonou, Benin Casamance, Senegal Libreville, Gabon Banjul, Gambia Bonny, Nigeria Caio, Guinea-Bissau Freetown, Sierra Leone Cachen, Guinea-Bissau Calabar, Nigeria Abidjan, Cte d'Ivoire Conakry, Guinea Monrovia, Liberia Luanda, Angola Babaque, Guinea-Bissau Port Etienne, Mauritania Spring Tidal Range (m) 0.80 0.92 0.97 1.23 1.26 1.30 1.40 1.50 1.68 1.82 2.30 2.58 2.74 3.07 3.10 3.50 3.60 4.20 4.24 5.80

Based on various sources including Lawson (1966), Diop (1990) and the National Tidal Facility, Flinders University, South Australia.

42

Table 2. Geomorphological and hydrological settings of mangroves along the Atlantic coast of Africa.
Deltaic --------------------------Estuarine Fluvial + + + + + + + + + + + + + + + + - absent + + + + + + + + + + +

Frontal Mauritania Senegal Gambia Guinea-Bissau Guinea Sierra Leone Liberia Cte d'Ivoire Ghana Togo Benin Nigeria Cameroon Equat. Guinea Gabon Congo Zaire Angola + + + + ? + + present

Lagoonal + + + + + + + + + ? -

? insufficient data

43

Table 3. Listing of major families (with genera and species) of mangroves and significant mangrove associates on the Atlantic coasts of America and Africa and the eastern coast of Africa.
Atlantic America Avicenniaceae Avicennia germinans A. schaueriana Rhizophoraceae Rhizophora mangle R. harrisonii R. racemosa Combretaceae Laguncularia racemosa Conocarpus erectus Meliaceae Carapa procera Carapa guyanensis Sterculiaceae Sonneratiaceae Malvaceae Hibiscus tiliaceus Thespesia populnea Rosaceae Chrysobalanus icaco Theaceae Pelliciera rhizophoreae Acrostichaceae Acrostichum aureum A. danaefolium Lythraceae Palmeae Mauritia flexuosa Raphia taedigera * Nypa fruticans Fabaceae Pterocarpus officinalis Drepanocarpus lunatus Dalbergia ecastaphyllum Mora oleifera Pandanaceae Aizoaceae Sesuvium portulacastrum Poaceae Sporobolus virginicus Atlantic Africa A. germinans R. mangle R. harrisonii R. racemosa L. racemosa C. erectus C. procera H. tiliaceus T. populnea Chrysobalanus ellipticus C. orbicularis A. aureum Phoenix reclinata P. canariensis Raphia hookeri *Nypa fruticans P. santalinoides D. lunatus D. ecastaphyllum Pandanus candelabrum S. portulacastrum S. virginicus * introduced species Indian Ocean Africa A. marina R. mucronata Bruguiera gymnorrhiza Ceriops tagal Lumnitzera racemosa Xylocarpus granatum X. moluccensis Heritiera littoralis Sonneratia alba H. tiliaceus T. populnea T. acutiloba A. aureum A. speciosum Pemphis acidula Barringtonia racemosa P. reclinata Dalbergia spp. S. portulacastrum S. virginicus

44

Table 4: Endemic mangrove species of each of the major mangrove regions of the World.
Region Pacific America: Atlantic America: West Asia: South-East Asia: North-East Asia: Australia & Pacific: Endemic Mangrove Species A. bicolor A. schaueriana Sonneratia apetala, Heritiera fomes A. alba, A. lanata, Bruguiera hainesii Sonneratia hainanensis, Kandelia kandel A. integra, Sonneratia lanceolata.

45

Table 5. Distribution of mangroves and associated species from the four islands in the Gulf of Guineaa
Speciesb A. germinans R. harrisonii A. aureum C. erectus H. tiliaceus C. procera D. lunatus P. candelabrum R. racemosa R. mangle L. racemosa P. reclinata Chrysobalanusd Area (km2) Max. Altitude (m) Dist. to mainland (km) Dist. to landward neighbour (km) Number of Species 1957; Fernando Po (= Bioko) + + + + + + + + + + + + 2,208 2,850 32 32 12 Principe +c + + + + 126 948 215 210 5 Sao Tom + + + + + + 964 2,024 235 135 6 Annobon 18 655 340 180 0

a based on data from Hutchinson et al., 1954/72; Exell, 1945; 1973; Keay, 1953; Adams, b Thespesia populnea has been omitted from this table as it has been widely planted in the c

towns on the islands; this unconfirmed record is based on a description for which no voucher specimen could be located (Exell, 1945); d includes two species (see Table 3) both of which are present on the adjacent mainland (Wouri Estuary, Cameroon) but not recorded from any of the islands.

46

Table 6. Distribution of the mangrove species and associates along the Atlantic coast of Africa.
Location Mauritania Senegal Gambia Guinea Guinea-Bissau Sierra Leone Liberia Cte d'Ivoire Ghana Togo Benin Nigeria Cameroon Eq. Guinea Gabon Congo Zaire Angola Av x x x x x x x x x x x x x r x x x x La x x x x x x r x x x r x Co r x x x x x x r x x x x x x x Rr r x x x x x x x x x x x x x x x x x Rm x x x x x x x x x x Rh x x x x x x x r x x x x Nf x x Aa x x x x x x x x x x x x Ht Tp Dl x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x Chr Pc x x

x x x x x x x x x

x present; r rare; - definitely absent; Compiled from numerous sources given individually under 'Regional Descriptions' below. Species abbreviations are as follows: Av Avicennia germinans; La Laguncularia racemosa; Co Conocarpus erectus; Rr Rhizophora racemosa; Rm Rhizophora mangle; Rh Rhizophora harrisonii; Nf Nypa fruticans; Aa Acrostichum aureum; Ht Hibiscus tiliaceus; Tp Thespesia populnea; Dl Drepanocarpus lanatus; Chr Chrysobalanus spp.; Pc Pandanus candelabrum.

47

Table 7. Upriver distances (km) reported for mangrove limits.

River Cacheu R., Guinea-Bissau Saloum R., Senegal Bonny R., Nigeria Zaire R., Zaire Wouri R., Cameroon Muni R., Equatorial Guinea Ntem R., Equatorial Guinea Distance upstream 100 70 70 50 40 17 15 Reference Sayer et al., 1992 Sayer et al., 1992 Dublin-Green, 1990 Delevoy, 1945 Boy et al., 1975 Sayer et al., 1992 Sayer et al., 1992

48

Table 8. Areal extent of mangrove vegetation along the Atlantic coast of Africa together with selected demographic and economic indicators.
Country Coastline Length (km) Mauritania Senegal Gambia Guinea-Bissau Guinea Sierra Leone Liberia Cte d'Ivoire Ghana Togo Benin Nigeria Cameroon Equatorial Guinea Gabon Congo Zaire Angola 754 531 80 350 320 402 579 515 539 56 121 853 402 296 885 169 37 1,600 Population Density (N/km2) 2.1 37.4 70.0 34.4 27.3 54.3 25.9 35.5 67.5 61.5 41.4 120.1 23.5 15.3 4.8 5.7 14.9 7.8 (N/km) 2,785 13,559 8,750 2,857 20,938 9,701 4,318 21,942 28,757 58,929 38,016 128,253 27,363 1,351 1,356 11,243 913,514 6,063 Mangrove Area (km2) 10 1,853 497 2,484 2,963 1,838 190 150 100 26 69 10,515 2,434 257 6,129 120 226 1,250 Per Capita GNP (US$) 500 710 260 180 480 240 440 730 390 410 360 270 940 330 3,220 1,010 230 620

Based on various data sources including Saenger et al., 1983; Hutchings and Saenger, 1987; Sayer et al., 1992; Diop et al., 1993.

49

Table 9. Structural analysis of plants >3m in height in forty-four 100m2 quadrats from south-eastern Nigeria. (From Upkong, 1992).
Species Frequency Rel. Density % freq. (stems/ha) % 63.0 41.4 44.0 24.4 20.0 14.8 5.0 5.0 4.0 28.4 18.7 19.9 11.0 9.0 6.8 2.2 2.2 1.8 100 200 98 68 40 40 25 8 8 5 492 Rel. dens. % 40.7 20.0 13.8 8.1 8.1 5.1 1.6 1.6 1.0 100 Cover % 28.6 10.4 10.0 3.8 5.0 2.3 0.5 0.5 0.5 61 Rel. cover % 46.4 16.9 16.3 6.2 8.1 3.7 0.8 0.8 0.8 100 I.V.

A. germinans R. racemosa R. mangle R. harrisonii N. fruticans P. reclinata P. candelabra Triumfetta rhomboideae Drepanocarpus lunatus Total

115.5 55.6 50.0 25.3 25.2 15.6 4.6 4.6 3.6 300

50

Table 10. Structural composition of the littoral vegetation in the delta of the Soumba River, Guinea (derived from analysis of SPOT image 30-330 taken on 19 April 1986).
Community Tall dense mangrove Medium dense mangrove Short dense mangrove Short sparse mangrove Saltflats & herblands Riceponds Total littoral vegetation Area (ha) 402.1 856.7 2269.2 1764.9 619.4 369.2 6281.5 % littoral vegetation 6.4 13.6 36.1 28.1 9.9 5.9 100

51

Table 11. Flowering times of the principal mangrove species at various localities on the Atlantic coast of Africa.
Location Senegal Gambia Guinea Sierra Leone Liberia Cte d'Ivoire Ghana Benin Nigeria F. Po/S. Tome Cameroon Gabon Congo Angola Av La 11 Co 12-11 Rr 1/56 Rm 25 Rh 15 56 56 3/6/10-116

47 57 3-63,4 12-67,8 1-22,

3-61 71 41 129 121 111 11-122

3/121

12-36 11-126 105/26 76 2-116,8 12-22,6 26 16 126

76

101 2-32

116

116 106 1/86

6/116

Sources: 1 Hutchinson et al. 1954/72; 2 Din 1991; 3 Blasco 1985; 4 Baglo, 1989; 5 Assemien 1969; 6 Keay 1953; 7 Saenger pers. obs.; 8 Jackson 1964; 9 Paradis 1979 Species abbreviations are as follows: Av Avicennia germinans; La Laguncularia racemosa; Co Conocarpus erectus; Rr Rhizophora racemosa; Rm Rhizophora mangle; Rh Rhizophora harrisonii.

52

Table 12. Monthly distribution of flowering times recorded for mangroves and associated species divided between dry and wet season. (For sources see table 11).
Species Avicennia germinans Rhizophora racemosa Rhizophora harrisonii Rhizophora mangle Laguncularia racemosa Conocarpus erectus N D J F M A + + + + + + + + + + + + + + + + + + + + + + + + + M J J A S O + + + + + + + + + + + + + +

53

Table 13: Seed-fall in mangroves and associated species at Ikoroda, Western Nigeria (from Jackson, 1964)
Species Avicennia germinans Rhizophora racemosa Hibiscus tiliaceus Drepanocarpus lunatus Dalbergia ecastaphyllum Pterocarpus santalinoides Ormocarpum verrucosum Cynometra megalophylla Seed-fall October - January February - March March June - September June - December June - August May - December May

54

Table 14. Monthly distribution of fruiting times recorded for mangroves and associated species divided between dry and wet season. (For sources see table 11).
Species Avicennia germinans Rhizophora racemosa Rhizophora harrisonii Rhizophora mangle Drepanocarpus lunatus Dalbergia ecastaphyllum Hibiscus tiliaceus Pterocarpus santalinoides Cynometra megalophylla N D J F M A + + + + + + + + + + + + M J J A S O + + + + + + + + + + + + + + + + + + + + +

55

Table 15. Structural data for the mangroves in the Wouri Estuary, Cameroon (from Din, 1991).
Species Stems/ha Basal Area (m2/ha) 8.66 5.59 0.07 Mean (cm) 18 31 8 Freq (%) 59.5 32.4 8.1 Dom (%) 60.5 39.0 0.5 Dens (%) 79.4 17.2 3.4 Importance value (%) 199.4 88.6 12.0

R. racemosa A. germinans D. lunatus

305 66 13

56

Table 16. Export data for Rhizophora timber from the Wouri Estuary, Cameroon between 1919-37.
Year 1919-26 1927 1928 1929 1930 1931 1932 1933 1934 1935 1936 1937 tonnes 30,000 5,400 no data 5,600 3,118 1,249 324 373 2,812 1,333 766 553

From Hdin (1928) and the League of Nations, Annual Mandate Administration Reports quoted by Grewe (1941).

57

Table 17. Potential acid sulphate soils in West Africa (from Marius, 1984).
Country Senegal Gambia Guinea-Bissau Guinea Sierra Leone Liberia Cte d'Ivoire Ghana-Togo Nigeria Cameroon km2 9,750 3,750 11,750 8,250 5,000 1,250 250 1,500 15,500 2,500

58

Table 18. Existing and proposed protected areas containing some mangrove vegetation on the African Atlantic coast.
Country EXISTING Mauritania Senegal Senegal Gambia Gambia Gambia Guinea-Bissau Guinea-Bissau Guinea-Bissau Guinea-Bissau Guinea-Bissau Sierra Leone Sierra Leone Sierra Leone Ghana Ghana Ghana Ghana Ghana Nigeria Gabon Gabon Gabon Congo Zaire Angola PROPOSED Guinea-Bissau Guinea-Bissau Guinea-Bissau Guinea-Bissau Liberia Sierra Leone Tarrafes NP Bijagos Archipelago Cacheu River Cantanhez Forest Cape Mont Nature Reserve Bunce River National Park status uncertain status uncertain status uncertain Nature Reserve status uncertain proposed proposed proposed proposed proposed proposed 286 na na na na na Parc National du Banc d'Aguin Saloum Delta NP Basse-Casamance NP Gambia River NP Kiang NP Niume/Sine Saloum NP Cufada Lagoon Iles Tristao Konkour Rio Kapatchez Rio Pongo Bonthe Mangrove Swamp Sulima Mangrove Swamp Bumpe Mangrove Swamp Anlo-Keta Lagoon Densu Delta Muni Lagoon Sakumo Lagoon Songor Lagoon Cross River NP Petit Loango Sett Cama Wongha-Wongh Conkouati Wildlife Reserve Zaire R. Mangrove Reserve Parque Nacional da Qicama National Park World Heritage Site National Park National Park National Park National Park National Park Ramsar Site Ramsar Site Ramsar Site Ramsar Site Ramsar Site Nature Reserve Nature Reserve Game Sanctuary Ramsar Site Ramsar Site Ramsar Site Ramsar Site Ramsar Site National Park Ramsar Site Ramsar Site Ramsar Site Nature Reserve Nature Reserve Game Reserve National Park 1976 1989 1976 1970 1978 1987 1986 1990 1992 1992 1992 1992 na na na 1992 1992 1992 1992 1992 1991 1986 1986 1986 1980 1992 1938 1957 11,730 760 50 25 110 49 391 850 900 200 300 101 26 49 1,278 46 87 13 287 4,000 4,800 2,200 3,800 1,443 1,000 9,960 Name Status Date Area (km2)

Based on various data sources including Saenger et al., 1983; Sayer et al., 1992; Diop et al., 1993; IUCN, 1994; and personal communications.

59

15

y = 2.0141 + 0.0045x r = 0.93

No. of Species

10

0 0 1000 2000 Area (sq. km) 3000

Fig. 5. Number of mangrove and associated species against the area of the Gulf of Guinea islands of Fernando Po, Principe, Sao Tom and Annobon.

60

15 y(M) = 13.52 - 0.038x r = 0.98

Number of Species

10

y(I) = 13.04 - 0.052x r = 0.83 0 10 100 1000

Distance(km)

Fig. 6. Number of mangrove and associated species against the distance from the mainland y(M) - and from the nearest landward island - y(I) - of the Gulf of Guinea islands of Fernando Po, Principe, Sao Tom and Annobon.

61

Nigeria Guinea Guinea-Bissau Gabon Gambia Zaire Cameroon Sierra Leone Senegal Eq. Guinea Angola Congo Benin Togo Liberia Ct e d'Ivoi re Ghana Mauritania

5 10 Mangrove Area/Coastline Length

15

Fig. 7. Mean breadth of mangrove vegetation in West Africa, calculated from estimates of total mangrove area and length of coastline.

62

Legends for Plates Plate 1. Dense stands of Rhizophora racemosa 10-12m high in the Wouri Estuary, Cameroon. Plate 2. Dense stands of Rhizophora racemosa 10-12m high, forming a fringe on the coastal lagoon near Grand Lahou, Cte d'Ivoire. Plate 3. Narrow waterway amongst the fringing Rhizophora racemosa of the coastal lagoon near Gran Lahou, Cte d'Ivoire. Plate 4. Low stands of Rhizophora mangle in the Bonny River near Port Harcourt, Nigeria. Plate 5. Mangrove vegetation of Sukoma Lagoon, near Tema, Ghana, dominated by Avicennia germinans up to 6m high. Plate 6. Korle Lagoon, near Accra, Ghana, showing the degree of infilling due to sewage and stormwater inputs. Despite the continuously open connection with the ocean, the mangroves are dominated by Avicennia germinans.